As a parent of a teen and another child who will be one soon, the subject hits home for me. We worry about the influence of social media on kids.

Some want to ban it for anyone under 16. I’m not against the ban. However, does controlling access solve the problem? When something is off-limits, we often want it more. Prohibition rarely works. Instead, it takes them to something more appealing where they get attention. This slowly becomes a problem, as when they don’t receive attention (likes), the real issue emerges. Phones or social media are not the only problem. Parenting is where it starts.

Follow on LinkedIn

The conversation often turns to restricting access. But do you think control is really the solution, or the first line of defense? The truth is, prohibition can often make something more appealing. I believe the real challenge isn't with the technology itself, but in how we parent.

We often give kids devices to occupy them during our own busy moments—at the gym, during dinner, or with friends. Over time, that device becomes their sanctuary.

We, as parents, will have to take the first responsibility and accountability. Instead of building walls, we need to create awareness. We need to be more involved in their lives. Listening to our children as much as possible, rather than telling them what to do, will go a long way.

This is a work in progress for my family, but I trust we're moving in the right direction by focusing on conversation and understanding rather than just rules.

What do you think? Are we ready to parent intentionally in the digital age?

Leave a comment

Binary Circuit investigates trends, technology, and how organizations might profit from rapid innovation. GreenLight, the brain behind Binary Circuit, finds possibilities, analyzes challenges, and develops plans to fit and grow firms to stay ahead. Looking to scale your business, partners, or AI/technology integration.

Schedule a discovery call

This week, we explore how organoids aid in studying diseases and treatments, and how increased access to AI supercomputing accelerates R&D in next-generation technologies.

Mini organs are on track to revolutionize modern medicine

All over the world, scientists are growing tiny versions of human organs from hearts and brains to livers and intestines. These are known as organoids, and now this field is rapidly advancing in pediatric heart care.

The Australian research team at QIMR Berghofer Medical Research Institute, along with collaborators from Murdoch Children's Research Institute and The Royal Children’s Hospital, has developed a method to transform immature heart cells derived from stem cells into lab-grown tissues that function like real, fully mature adult heart muscle.

To "train" the cells into maturity, the researchers activated two biological pathways known to promote heart development: AMPK and ERR. The results are revolutionary for cardiac research and were published in Nature Cardiovascular Research. Conditions brought on by mutations in important genes such as CASQ2, RYR2, and desmoplakin—genetic abnormalities known to cause major heart problems in children—were successfully reproduced by the team.

These lab-grown hearts now offer a powerful, non-invasive model for scientists to:

• Understand how specific mutations cause heart problems

• Test multiple drugs on tissues created from a single patient’s own cells

• Accelerate personalized therapies—possibly avoiding risky surgeries or transplant

In fact, in one trial, scientists reversed disease damage in organoids using a new class of drugs called BET inhibitors—a big step toward real-world treatments for desmoplakin cardiomyopathy, a severe and often deadly condition.

Challenges still persist but organoid research is progressing faster than ever before. Stem cell science, 3D printing and better lab techniques together are making it easier to build human organs in the lab.

AI supercomputing is becoming more accessible

Supercomputers were once exclusive, limited to government labs and research centers. However, this is rapidly shifting as their accessibility improves.

AI has transformed computation studies, shifting from desktop workstations to continent-scale infrastructure, making high-performance AI computing more accessible and vital.

In July, Georgia Tech was granted $20 million by the U.S. National Science Foundation to develop Nexus, an AI supercomputer intended for public use. Nexus aims to accelerate the progress in drug discovery, robotics, among other research areas by providing computing power capable of processing over 400 quadrillion operations/sec. Nexus will be open to researchers nationwide and is expected to be completed by spring 2026.

Across the Atlantic, the UK launched Isambard-AI, a zero-carbon supercomputer with over 5,400 NVIDIA GH200 chips. Isambard AI is the UK’s top supercomputer, with the lowest carbon footprint, operating at 21 exaFLOPS. It ranks 11th globally and fourth on the Green500 list for energy efficiency, demonstrating progress in making AI machines and data centers more environmentally friendly.

Accessibility over speed sets it apart. It supports multimodal language models, such as Nightingale AI and BritLLM, and is open to the research community.

In Asia, Foxconn partners with U.S. chipmaker Nvidia to create a next-gen "AI Factory," a supercomputing center in Taiwan. With over 10,000 Blackwell GPUs, it will support industries from manufacturing to smart technology infrastructure.

NVLink Fusion enables custom chips and third-party CPUs to connect directly to Blackwell systems, allowing organizations to build efficient AI supercomputers. Notably, industrial AI growth, at 2.7x, is outperforming the public sector’s 1.9x growth by a significant margin. As deployments increase, the industry's portion of global AI compute has grown from 40% in 2019 to 80% 2025.

These advancements show that AI supercomputing is rapidly evolving from a niche technology to an essential public utility for the nation's progress.

Chart of the week:

Solar becomes the top contributor to the EU's power mix

June was a monumental month in the European Union: For the first time, it got more electricity from solar power than from any other source.

New data from energy think tank Ember shows that solar accounted for 22.1% of the EU’s electricity mix in June 2025, narrowly overtaking nuclear – and, notably, far outpacing gas and coal combined.

At least 13 EU countries set new records for solar generation in the last month. It’s a striking illustration of how far solar power and clean energy have come in the EU.

While the USA seeks to tether the U.S. to fossil fuels, Europe and much of the world continue accelerating toward cleaner options.

Binary Circuit investigates trends, technology, and how organizations might profit from rapid innovation. GreenLight, the brain behind Binary Circuit, finds possibilities, analyzes challenges, and develops plans to fit and grow firms to stay ahead. Looking to scale your business, partners, or AI/technology integration.

Schedule a discovery call

This week, we discuss AI’s role in reducing emissions is underestimated. Gene therapy brings big promises and dangers, and robots are becoming part of everyday life quicker than expected.

AI’s role in reducing emissions is underestimated

Artificial intelligence (AI) is power hungry, yet could also help in managing climate change by optimizing various economic sectors. A recent study in Nature magazine concluded that AI’s efficiency optimization across just three economic sectors could reduce worldwide emissions by up to 5.4 billion tonnes annually, which is more than the total emissions of the United States.

The “Green and intelligent: the role of AI in the climate transition” report, authored by the London School of Economics and Systemiq, shows that AI applications in food, electricity, and mobility could unlock enormous environmental benefits.

This saving exceeds estimated increases in emissions from AI-driven data centre energy use, indicating a strong net-positive climate contribution.

AI’s impact - key statistics

• Food: AI can reduce sectoral emissions by 0.9 to 1.6 billion tonnes of carbon annually in a "business as usual" (BAU) scenario to 18–33% in an ambitious AI scenario

• Energy (Electricity): AI can optimize grid management and increase solar PV and wind load factors by 20% in the power sector. This might cut carbon emissions by 1.8 Gt per year by 2035

• Mobility (Transport): Expected emission cut of 0.5–0.6 billion tonnes of carbon per year. AI-enhanced shared mobility improves vehicle utilization, cutting total kilometres driven, and reduces carbon footprint

Accelerating resource efficiency (Use Cases):

• The National AI Strategy of Singapore and Google DeepMind's wind optimisation tool, which raised wind energy value by 20%, are prime examples of AI's efficacy in this field.

• AI is driving the innovation, with tools like DeepMind's GNoME identifying 2 million new materials that could revolutionise renewable energy storage.

• AI also improves efficiency and reduces waste in industrial sectors. Amazon's AI-powered packaging optimization has saved more than 3 million tonnes of material since 2015.

AI is arguably one of the world’s most powerful climate allies. AI’s smart applications and real-time insights can help optimize many more sectors that contribute toward global warming.

Gene Therapy- big hopes or real risks. What comes next?

Gene therapy is moving quickly toward becoming part of everyday medicine, it now faces more pressure to prove it's safe and reliable. As per Gene, Cell, & RNA Therapy Landscape Report Q2 2025, 31 gene therapies have been approved (including genetically modified cell therapies).

Researchers in Australia have made a significant breakthrough with CAR-T cell therapy, which has so far been effective primarily for blood cancers. Using a tool called CRISPR, they reprogrammed immune cells to target solid tumors like breast and colon cancer, without harming healthy tissue. In animal tests, nearly every case was cured. If human trials go well, this could be beneficial for cancer treatment.

Scientists have successfully restored hearing in all 10 participants of a new clinical trial—some of whom were completely deaf since birth. The breakthrough, published in Nature Medicine, shows that a one-time gene therapy can reverse deafness caused by a mutation in a single gene. Before this treatment, most patients with OTOF mutations would need cochlear implants.

Biogen’s once-a-year shot for spinal muscular atrophy (SMA) is showing promise, and a skin disease treatment called Vyjuvek has been approved in Europe. These findings indicate that gene therapies are becoming more accessible for patients to use.

However, a gene therapy called Elevidys, used for children with Duchenne muscular dystrophy, was linked to two patient deaths. The company stopped treatment in some cases. Still, the future holds great promise. Eli Lilly acquired Verve Therapeutics for over $1 billion, as the company is developing a one-time gene editing treatment to lower cholesterol and reduce the risk of heart disease.

At the same time, the U.S. FDA is expected to tighten its approval rules. That means companies may need longer, more careful trials to prove their treatments are safe.

Looking ahead, the next 5 years point towards safer, smarter, and more affordable gene therapies. In short, gene therapy is entering a new phase. We believe potential is immense, but caution is warranted.

The robotic future is arriving faster than you think

The world is debating whether robots will replace human jobs but a transformation is soon to be seen both at our workplaces and homes.

Venture capitalist Vinod Khosla believes we’re just a few years away from a "ChatGPT moment" in robotics. By 2027–2028, he predicts, humanoid robots will move beyond rigid programming to become adaptive, learning machines capable of cooking meals, cleaning homes, and assisting in daily routines. By the 2030s, these machines will become as common as smartphones, with monthly rentals available for $300–$400.

Yet even as this vision takes shape for homes, we’re already witnessing its industrial reality most visibly at Amazon. The company recently deployed its one-millionth warehouse robot in a Japanese fulfillment center, nearly matching the number of human workers at its global facilities. Robots now assist in 75% of Amazon’s global deliveries, handling tasks like sorting, packaging, and transporting goods. These machines are driving productivity: from shipping 175 packages per U.S. employee annually in 2015 to nearly 3,870 today. Interestingly, the average number of human employees per facility is now at its lowest point in 16 years.

But instead of replacing humans, Amazon is changing the way we work. Over 700,000 employees have been retrained to manage or work alongside robots, as new hybrid roles emerge that blend machine oversight with human decision-making. Meanwhile, AI is enhancing everything from inventory placement to forecasting, and the company is already testing voice-command humanoid robots for tasks like trailer unloading.

These developments suggest that the robotic age isn’t too far—it’s a reachable reality. The Amazon warehouse today may foreshadow your home tomorrow. What began as task-specific industrial automation is rapidly evolving into general-purpose, learning-enabled assistants.

If Khosla’s forecast holds, the next frontier isn’t whether robots will enter our lives—it’s how seamlessly we’ll integrate them.

Advanced chipmaking capacity to see 69% growth through 2028

Global semiconductor manufacturing capacity is set to grow at a 7% CAGR from 2024 through 2028, reaching 11.1 million wafers per month (wpm), driven largely by AI adoption, as per SEMI’s 300mm Fab Outlook.

Advanced nodes (7nm and below) are expected to expand 69% over the same period, from 850,000 to 1.4 million wpm, growing at twice the industry average. Capacity for 2nm and below is projected to show even more aggressive scaling from under 200,000 wpm in 2025 to over 500,000 by 2028, as chipmakers race to meet AI-related demand.

Capital spending is also set to rise sharply.

Equipment investment for advanced process equipment is predicted to jump 94% to over $50 billion by 2028, while 2nm and below will see funding more than double to $43 billion—a 120% surge reflecting aggressive bets on next-gen technology.

A remarkable 120% increase explains the industry’s aggressive pursuit of next-generation manufacturing capabilities.

Binary Circuit investigates trends, technology, and how organizations might profit from rapid innovation. GreenLight, the brain behind Binary Circuit, finds possibilities, analyzes challenges, and develops plans to fit and grow firms to stay ahead. Looking to scale your business, partners, or AI/technology integration.

Schedule a discovery call

This week, we discuss as the race intensifies to define the future of energy, China’s AI explosion signals transformative global shifts, from medical superintelligence acting as digital twins of doctors to the surging consumer adoption of AI driven by increasingly complex life stages.

The race is on to define the future of energy

From nuclear plants in Turkey to battery farms in Saudi Arabia, China isn’t just exporting clean tech — it’s exporting long-term influence.

• $168B+ in clean energy foreign investments since 2023

• Africa’s largest wind farm is built by Chinese firms in Kenya

• EV production in Indonesia, nuclear in Pakistan, and solar in the Middle East. A lithium-ion battery plant by an Indonesian company and China's CATL is expected to be in operation by the end of 2026

These projects are backed by Chinese capital, workforce, and technology — laying the groundwork for enduring political and economic dependencies.

China added 57% of global wind and solar capacity in 2024. This isn’t just about green ambition, it is about global positioning.

U.S. Strategy: Fossil fuel leverage over long-term vision?

In contrast, the United States is prioritizing fossil fuel diplomacy.

• President Trump is pressuring allies like Japan and South Korea to invest “trillions” in U.S. LNG infrastructure.

• General Motors canceled plans for an EV motor plant in New York, redirecting $888 million into building V-8 gasoline engines instead.

• Dismantling foreign clean energy aid

• Pushing fossil exports (e.g. Alaskan gas to Asia)

• Linking gas sales to strategic alliances (e.g. Ukraine)

In the short term, this supports United States traditional energy customers.

But by midcentury, the IEA expects fossil fuels to fall below 60% of global energy demand — shifting long-term power toward clean tech suppliers.

Inside China’s AI Explosion: What It Means for the World

At the World Economic Forum in Tianjin, former PBOC official Zhu Min made a bold claim:
China is on track to produce over 100 DeepSeek-level AI breakthroughs in just 18 months.

The DeepSeek model is a real world example. Fueled by a massive, tech-literate population, world-class engineers, and state-backed R&D, China’s AI ecosystem has a unique structure. Even with U.S. semiconductor restrictions, domestic giants like Huawei are building in-house chips and compute power. High-tech already makes up 15% of China’s GDP, and it’s projected to hit 18% in two years.

Researchers at Harvard University also found China has advantages in two key building blocks of AI, data and human capital, that are helping it keep pace.

This isn’t just about AI models—it’s about economic reprogramming.
China is betting on AI to transform manufacturing, finance, healthcare, and education. Think of it less as “chasing GPT-4” and more like building a self-sustaining AI industrial base. Every breakthrough will ripple through the economy like a second industrial revolution.

So the question is no longer if China will accelerate—it’s how the world plans to respond.
Because if 100+ AI breakthroughs land by end-2026, we’re not just in a tech race we’re in a global economic reset.

Medical Superintelligence as Doctors’ Digital Twin: The Next Leap in Healthcare

Monthly breakthroughs in AI, each offering improved features, have become the new norm. Recently, Microsoft introduced a medical AI system that outperforms doctors on complex diagnostic cases. But can medical superintelligence actually match a human doctor’s diagnostic ability?

MAI-DxO (Medical AI Diagnostic Orchestrator) doesn’t rely on just one model. Instead, it uses a “chain-of-debate” approach where different AI models including OpenAI’s GPT, Google Gemini, Meta’s LLaMA, Anthropic Claude, and Elon Musk’s xAI debate and build on each other’s reasoning before offering a final diagnosis.

Based on real patient case studies from the New England Journal of Medicine MAI-DxO got it right 85.5% of the time. That’s four times better than doctors working without any decision-support tools. It also managed to cut unnecessary testing by nearly 20%.

However, real-life diagnosis isn’t always about textbook symptoms and clean data. Patients sometimes don’t describe things clearly. Medical histories can be incomplete. Emotions, fears, and social context all play a role. That’s where doctors excel and AI still lags behind.

Additionally, companies like Lila Sciences are building autonomous labs where AI doesn’t just diagnose its designs and runs experiments, generating new therapies without human input. Alibaba's GRAPE AI detects early-stage stomach cancer using simple CT scans, avoiding the need for invasive endoscopies.

Recursion Pharmaceuticals and Insilico Medicine are using AI for quick drug discovery in less time. Medical superintelligence doesn’t need to replace doctors. Real value lies in serving as a second opinion. If implemented, it can be highly beneficial in rural areas where clinics are left without expert support, or lack of specialists.

Can patients trust a medical AI system ?

Technically, yes in a few cases. But context, trust, and human judgment are still essential.

Who would you trust more for a diagnosis—AI or a human doctor?

Comparison of AI powered diagnostic agents by accuracy and average diagnostic test cost per case. Top performing agents appear toward the top left quadrant, reflecting higher accuracy and lower cost. The lower dotted line represents the performance range of the best individual foundation models. The purple line traces the performance of MAI-DxO across different configurations. The red cross indicates the average performance of 21 practicing physicians.

Consumer AI adoption is being driven by life-stage complexity

When we think about who’s adopting AI, we often look at demographics like age or income. But new data reveals something far more strategic.

Parents, for instance, are emerging as some of the most engaged AI users. Why? Because raising kids creates friction — endless decisions, multitasking, scheduling chaos — and AI tools that reduce this cognitive load are becoming indispensable. The pattern is clear: the more complex life becomes, the more useful AI feels.

Here’s what the numbers reveal:

This represents a subtle but important shift in how we should think about the AI adoption curve.

Business Implications:

• Companies must target high-friction life moments such as new parenthood, caregiving, or even moving homes, not just high-tech users

• Design AI tools that solve real-life pain points with minimal effort from users

Chart of the Week:

Critical minerals converting from commodity to geopolitical currency

Past week developments have revealed a decisive shift in the global critical minerals sector. As critical minerals are now geopolitical assets, central to energy security, clean tech, and even AI infrastructure.

The Democratic Republic of Congo (DRC) extended its cobalt export suspension, while Zimbabwe announced a 2027 ban on lithium concentrate exports, both efforts to force local value addition. At the same time, China continues to control 70%+ of rare earth and graphite supplies, triggering renewed urgency among Western nations.

The U.S., EU, and G7 are responding with strategic countermeasures: stockpiling, direct investments in allied nations, and aggressive policy moves like the U.S. Section 232 probes into processed minerals and the EU’s Critical Raw Materials Act.

Brazil, South Africa, and Australia are stepping up to fill supply chain gaps.

Interestingly, even as production rises—lithium up 16%, cobalt supply from DRC hitting 44% growth—global prices dropped 16% last year. Yet this price decline hasn’t affected investment.

The race is no longer about price. It’s about control.

Binary Circuit investigates trends, technology, and how organizations might profit from rapid innovation. GreenLight, the brain behind Binary Circuit, finds possibilities, analyzes challenges, and develops plans to fit and grow firms to stay ahead. Looking to scale your business, partners, or AI/technology integration.

Schedule a discovery call

This week, we discuss AI era will be defined by human choices and not by code; AI advances cancer care through early intervention and sparks a new race centered on efficiency and environmental responsibility rather than scale.

The Era of AI Will Be Shaped Not by Code, but by the Decisions We Make

As AI rapidly enters into global economies, the 2025 Human Development Report talks about a shift that signifies AI not as a technological event, but as a policy and values-driven inflection point.

Global HDI (Human Development Index) reached its highest level ever in 2024, yet the rate of progress reached lowest in 35 years. At the same time, the gap between very high and low HDI countries is widening. The reason is not a failure of innovation, but a failure to align innovation with inclusive AI growth.

Tech is growing fast, but without proper systems, it deepens inequality. A global survey shows around 20% are currently using AI, and nearly two-thirds expect to use it soon—especially in education, health, and work across all HDI levels.

AI as an amplifier, not an equalizer

This report introduces the concept of a complementarity economy, where AI is used to augment not replace human potential. In radiology, for instance, AI enhances diagnostics, yet human judgment remains irreplaceable. In call centers, AI tools accelerate training but don’t diminish the need for empathy or problem-solving. But this promise is not distributed equally.

The majority of AI development both in model training and market alignment is happening in and for very high HDI countries. Meanwhile, low HDI countries remain data-scarce, infrastructure-poor, and excluded from global AI governance frameworks.

This asymmetry is also seen within countries. Even in high-income economies, men are more likely than women to use generative AI in professional settings, despite similar educational credentials. And in these same contexts, youth mental well-being is deteriorating, in part due to hyper-digitization and overexposure.

Implications for investors and policymakers

• Geoeconomic risk: AI concentration risks deepening digital inequality, leaving many nations dependent and excluded

• Workforce imbalance: Without human-centered design, AI may fuel underemployment and social instability

• Mental health strain: Rising digital stress among youth threatens future productivity, especially in aging economies

• Policy reset: Innovation must be judged not just by speed or scale, but by its real impact on human development

The coming years offer two paths:
- A tech future governed by code, scale, and capital concentration; or
- A human-centered future defined by inclusivity, ethical design, and institutional adaptability

The real risk isn't AI, but weak moral and political decisions. The future of AI will mirror the choices we make.

AI Is Transforming Cancer Care by Enabling Earlier Risk Detection and Preventive Action

Early detection in cancer has been limited by human perception and risk modelling for a long time. Doctors rely on radiology tests and other statistical frameworks found in family history, age or genetic predisposition. But sometimes the diagnosis is not timely or missed completely..

FDA’s recent approval of Clairity Breast, the first AI tool to predict a woman’s five-year breast cancer risk from a routine mammogram, is more than a regulatory milestone. It’s a signal that AI is moving upstream—from diagnostic support to predictive guidance.

Unlike traditional models, Clairity’s algorithm detects subtle, high-dimensional patterns in imaging data. Shift marks the early stages of a strategic realignment in healthcare logic. Instead of responding to disease, systems are beginning to pre-position around probabilistic risk—pushed forward by AI models trained on vast datasets and able to detect risk trajectories early.

Few developments include:

• Dxcover’s AI-powered multiomic cancer test provides early detection across multiple cancers through a single blood sample

• Garvan Institute’s AAnet model differentiates and characterises tumor cell subtypes in breast cancer, supporting multi-targeted treatment

• City of Hope’s generative AI platform integrates across clinical decision layers—from trial matching to operational planning

Collectively, these tools demonstrate a trend: AI is not simply assisting clinicians—it’s building new layers of anticipatory intelligence into cancer detection.

Points to consider:

• Risk is no longer one-size-fits-all; it’s personal, data-driven, and image-based

• AI can scan thousands, spotting diseased states before symptoms appear

• If doctors can’t see what AI sees, explainability becomes non-negotiable

• Predictive AI could spark action before diagnosis, blurring clinical lines

New AI race is not about scale - it’s about efficiency and emissions

For years, the generative AI race has been driven by a single assumption: bigger is better. Larger models, more parameters, and billion-dollar training runs became the norm. But this approach is now being questioned, as it comes with high costs and growing environmental concerns.

MiniMax-M1, a Chinese-developed open-source model that delivers performance on par with closed-source giants—at a training cost of just $534,700. With support for 1 million input tokens, state-of-the-art reasoning in mathematics and code generation, and a slim compute footprint, MiniMax-M1 has rewritten the cost-performance equation. In contrast, comparable models like DeepSeek R1 were trained at a cost of $5.6 million, and GPT-4 reportedly exceeded $100 million. In effect, MiniMax is showing that architectural intelligence now rivals parameter scale as a competitive differentiator.

MiniMax-M1 consumes only 25% of the floating point operations (FLOPs) required by DeepSeek R1 at a generation length of 100,000 tokens.
MiniMax-M1 outperformed closed-source competitors in mathematical reasoning and code generation, scoring 86% on AIME 2024.

Another dimension to this shift is environmental cost. A comprehensive benchmark analysis of 1,000-question performance evaluations reveals the steep energy and emissions impact of reasoning-focused large language models. For example, Deepseek-R1 70B delivered 78.9% accuracy—yet emitted 2,042.4 g CO₂eq per run, a 74x increase compared to smaller models like Qwen 7B. Even high-performing reasoning-enhanced models, such as Cogito 70B, while achieving 84.9% accuracy, generated large token volumes and emissions—pointing to a growing tension between intelligence and sustainability.

Combined CO₂ eq emissions (measured in grams) and overall accuracy of each LLM across 1,000 questions. The results illustrate trade-offs between model size, reasoning depth, and environmental impact.

As AI systems are deployed at scale—in education, healthcare, legal reasoning, and scientific discovery—token generation and response verbosity become environmental liabilities. Cogito 8B generated over 37,000 tokens on a single abstract algebra question. In contrast, Qwen 2.5 72B matched 77.6% accuracy in multiple-choice tests with just 1.0 word per answer, consuming a fraction of the energy.

The AI race is no longer just a competition of capability—it is becoming a test of constraint. The winners will be those who can achieve high reasoning performance under tight compute and emissions budgets.

Chart of the week:

Carbon pricing crosses $100B—But is it enough to catch up with rising CO₂?

For thousands of years, Earth’s CO₂ levels stayed stable at ~280 ppm. Today, they’ve surged past 430 ppm, a 50%+ increase driven largely by fossil fuel combustion. The science is clear, and the urgency is real.

In response, governments are finally putting a price on carbon.
According to the World Bank’s 2025 report, carbon pricing tools like taxes and emissions trading systems generated $100+ billion in 2024 - up from just a handful of systems in 2005 to 80 in operation globally today.

These mechanisms now cover 28% of global emissions, signaling growing momentum toward market-driven climate action. But-

Transport and agriculture—major emission sources—remain largely unpriced. CO₂ levels are still climbing, and so is climate volatility.

The question isn’t whether carbon pricing works—it’s whether we’re scaling it fast and wide enough. Cutting energy waste, switching to renewables, and preserving forests are no longer optional they’re essential for any credible climate strategy.

Sound Bite:

• Novo Nordisk plans to launch its blockbuster drug Wegovy in India, aiming for ₹8,600 crore in sales over 5–7 years. With its “20-20” strategy focused on obesity and heart health, Novo is eyeing the top spot in India’s prescription drug market

• Amazon will invest $233M+ in 2025 to expand its Indian logistics network. Expect faster fulfillment, improved infrastructure, and service coverage to every PIN code in the country

• UAE invites AI to the cabinet table. The UAE’s National AI System will become an official Cabinet advisor in 2025, also joining the Ministerial Development Council and federal board meetings

• Japan is building FugakuNEXT, a next-gen supercomputer projected to be 1,000× faster than existing systems. Powered by FUJITSU-MONAKA3 CPUs, it aims to lead in AI and scientific simulations

Binary Circuit investigates trends, technology, and how organizations might profit from rapid innovation. GreenLight, the brain behind Binary Circuit, finds possibilities, analyzes challenges, and develops plans to fit and grow firms to stay ahead. Looking to scale your business, partners, or AI/technology integration.

Schedule a discovery call

This week, we discuss by enabling China to draw in top scientific minds through shifts in U.S. policies, a global talent reshuffling is unfolding and also how AI-driven weather forecasting begins to gain significant momentum.

How U.S. policies are helping China to attract the best scientific talent?

As the U.S. becomes less hospitable to international talent and restricts access to its research operations, China is offering an attractive alternative.

In February, the Trump administration cut over $1.5 billion from the NIH budget, putting critical research at risk. Chinese scientists were also labeled as security risks.

China, by contrast, has been proactive.

Between 2010 and 2021, nearly 12,500 scientists of Chinese descent left the United States for China, with more than half departing between 2017 and 2021 alone. Once, nearly 87% of Chinese STEM PhDs in the U.S. stayed on; now, an increasing number are returning—lured by the promise of generous funding and rising prestige at institutions like Tsinghua, Zhejiang, and Westlake University.

This reversal of brain drain could spark a new era of global talent wars not just for talent, but for talent retention.

China has turned U.S. policy into a strategic advantage.

Local governments are offering generous financial incentives and benefits to attract researchers. For example: Gulin County (Sichuan) offers PhD holders a lump sum of 300,000 yuan and a monthly stipend. Taizhou (Zhejiang) provides up to 100,000 yuan for university graduates to relocate. Hunan province offers up to 1 million yuan to doctoral graduates returning from abroad.

According to the OECD, China’s gross domestic expenditure on R&D (GERD) reached $780.7 billion in 2023, nearly equal to the U.S. total of $823.4 billion. With China growing at 8.7% annually and the U.S. lagging at 1.7%, China is poised to take the global lead in real terms.

China’s domestic talent base is maturing.

Companies like DeepSeek, which built advanced AI models using teams from Chinese universities, highlight a growing confidence that world-class innovation can emerge from within. The old narrative where “foreign” equaled superior is fading. If China continues to offer competitive incentives, stable funding, and scientific autonomy, it could become a top notch center for scientific talent.

Will the U.S. rethink their R&D policies to retain and attract top scientific talent? Or will they allow geopolitical anxieties and fiscal shortfalls to erode their scientific edge?

AI-driven weather forecasting is picking up momentum

Weather forecasting has seen meaningful progress over the last few decades. For example, a five-day forecast now is as reliable as a three-day forecast was in 2000. In the U.S. alone, an estimated one-third of GDP—roughly $3 trillion—is sensitive to weather variability.

Earlier weather prediction is based on physics-based numerical models powered by supercomputers. Key global forecasting systems include the U.S. Global Forecast System (GFS) and Europe’s ECMWF model. While the ECMWF is generally regarded as more accurate due to its finer resolution (9 km vs. GFS’s 13 km), it also requires greater computing power and is not as widely accessible as the GFS.

As climate volatility intensifies, the demand for timely, localized, and cost-effective forecasts is expected to grow. However, the sector is undergoing structural disruption, driven by advances in artificial intelligence.

Aardvark Weather, an AI-based system from the University of Cambridge, supported by the Alan Turing Institute, Microsoft, and ECMWF, outperforms the U.S. GFS model on several metrics using just 10% of the input data.

Major technology firms are also entering the space. Microsoft developed a weather forecasting model that makes its predictions using artificial intelligence. Google, Nvidia and Huawei are also doing the same. Google DeepMind has introduced GraphCast, an AI model now being used operationally by government forecasters. Recently they have launched Weather Lab, an interactive platform showcasing their AI weather models and is capable of tracking hurricanes up to 15 days in advance predicting both trajectory and intensity.

AI also addresses longstanding pain points in meteorology. Traditional models struggle with hyper-local forecasts due to resolution limits and computational constraints. Nvidia’s CorrDiff can upscale coarse simulations into high-resolution outputs with reduced energy and less time.

Similarly, rainfall prediction in the tropics was not easy due to limited observational infrastructure and complex atmospheric dynamics but it’s seeing improvements through AI.

ECMWF, the Hong Kong Observatory, and startups like Ignitia are deploying machine learning to improve short-term precipitation forecasts in equatorial regions.

Subseasonal forecasts are critical for agriculture, disaster preparedness, and energy markets but suffer predictability issues. A model developed by Fudan University helps in predicting the Madden-Julian Oscillation, a key atmospheric driver of weather variability.

AI is catalyzing a foundational transformation in weather forecasting. As accuracy improves and costs decline, AI-powered forecasting could become a key enabler of resilience and efficiency in an increasingly volatile climate economy.

Chart of the week

Clean energy investment in 2025 will be twice as much money going to fossil fuels: IEA

Despite global uncertainty, energy investment is set to hit a record $3.3 trillion in 2025. Clean energy is taking the lead with twice the amount going to fossil fuels ie., ~$2.2 trillion. Solar falls in the largest single investment category.
China is by far the largest energy investor globally, spending almost as much as the EU and United States combined. Over the past decade, the country's share of global clean energy investment has increased from 25% to nearly 33%, underpinned by strategic investments in solar, wind, hydropower, nuclear, battery storage and electric vehicles.

Binary Circuit investigates trends, technology, and how organizations might profit from rapid innovation. GreenLight, the brain behind Binary Circuit, finds possibilities, analyzes challenges, and develops plans to fit and grow firms to stay ahead. Looking to scale your business, partners, or AI/technology integration.

AI’s massive potential will only be realized over time.

The promise of AI, especially generative AI (genAI), is immense, as reflected in trillions of dollars in projected economic gains over the next few years. At the same time, evidence is rising that AI project failures are increasing, and recent pullbacks in infrastructure investments, mainly in data centers, raise questions about whether the buildout is an overkill. Productivity gains and transformative impacts have not fully materialized in aggregate statistics.

This is a classic pattern of a general-purpose technology. Jeffrey Ding, a Professor at George Washington University, argues in his recently published book, Technology and the Rise of Great Powers: How Diffusion Shapes Economic Competition, that GPTs don't transform economies overnight. Their real power comes not from early innovation but from long-term diffusion. It requires continuous complementary innovation to integrate GPT across other industries, upskill the workforce, and support infrastructure development and government policies.

However, unlike previous GPTs, GenAI is the first technology in human history that has become accessible to such a large population almost overnight. Furthermore, a natural language processing web interface has made it possible for everyone to use it, creating unrealistic expectations. This anxiety leads to the “trough of disillusionment,” where initial euphoria turns to disappointment as implementation challenges arise.

A recent Economist article notes that, according to Gartner’s Research Vice President John-David Lovelock, AI’s “trough of disillusionment” could last until the end of next year. This will be followed by the “slope of enlightenment,” under which companies that have underinvested in GenAI start to regret it. The cost of falling behind in GenAI integration is higher than any other technology.

AI could transform business dynamics through new revenue sources and drive productivity overnight is exaggerated. Instead, the technology’s benefits will materialize over the next few years, though perhaps sooner than previous GPTs.

In this regard, economist Erik Brynjolfsson, who has long studied the intersection of technology and economics, stated in a podcast with Paul Krugman that GPTs typically take longer to demonstrate measurable impact because they necessitate considerable time and effort to transform business processes, reorganize companies, and reskill workers before their full productivity impact is realized.

This explains the recent increase in abandoned genAI projects. S&P Business Intelligence noted that the project failure rate has increased. The ratio of companies abandoning GenAI initiatives has risen from 17% to 42%, with 46% of proof-of-concept projects failing to reach production. The below chart shows the response to the question, “To the best of your knowledge, what percentage of AI projects in proof of concept at your organization are abandoned before production?”

A RAND study cites estimates suggesting that roughly 80% of the AI projects fail to achieve stated objectives, 2x the failure rate of broader IT projects. The study identifies key issues like a lack of high-quality data, an emphasis on cutting-edge technology rather than real-world problems, insufficient infrastructure investment, and applying AI to challenges beyond its current capabilities.

Businesses often fall into the “AI implementation trap” by adding new tools without reevaluating processes. This results in AI becoming an extra layer of effort. When combined with the explosion of disconnected AI tools and underappreciated learning curves, this can cause disappointment and early tool abandonment.

Brynjolfsson highlights the difficulty of assessing digital goods’ economic impact. Although services like search engines and social media have great consumer value, their zero price keeps them off traditional GDP. Furthermore, funding new technologies like genAI sometimes lowers first productivity figures. Companies invest in intangible assets like business process changes and training, which have expenses but are not entirely shown in GDP growth figures. Brynjolfsson’s team at Stanford created the GDP-B framework, a measure of these digital gains, noting their unmeasured worth of trillions of dollars in economic gains.

Brynjolfsson noted that adopting GPTs like genAI often follows a “Productivity J-Curve.” Productivity may initially dip as companies invest in reorganizing, retraining their workforce, and developing new processes, which creates unmeasured intangible assets. Productivity surges over time once complementary innovations and process changes catch up.

However, in due course, like all other GPTs, genAI will become invisible but indispensable. Like electricity, where value is seen through enhanced functionality and convenience rather than direct interaction, embedding AI into every good and service we use is accelerating for a seamless user experience.

With their vast ecosystems and billions of consumers, big-tech companies are including genAI in their primary operations and turning it into a utility that consumers interact with. Among notable instances are Google's generative AI in advertising and “AI mode” for search, which offers LLM-generated summaries. Meta has incorporated AI into advertisements, Microsoft has included it in applications like GitHub and Copilot, and Amazon utilizes it for recommendations and logistics.

As technology advances, it may fade into everyday life. In the next five to ten years, we may stop using the term “AI,” just as we no longer refer to “electrified factories” or “internet-enabled businesses.” Instead, these features are assumed and become a new normal.

Binary Circuit investigates trends, technology, and how organizations might profit from rapid innovation. GreenLight, the brain behind Binary Circuit, finds possibilities, analyzes challenges, and develops plans to fit and grow firms to stay ahead. Looking to scale your business, partners, or AI/technology integration.

Schedule a discovery call

This week, we discuss climate change’s threat to global public health, the rising energy demands of AI straining electric grids, the emergence of humanoid robots addressing labor shortages in manufacturing, and China's AI chip advancements challenging U.S. dominance.

Climate change poses an increasing threat to global public health and, in turn, to businesses’ growth

The WMO Global Annual to Decadal Climate Update (2025–2029) says global temperatures will stay close to record highs for the next five years. This will have a significant effect on businesses and public health.

The report says that from 2025 to 2029, the global near-surface temperature will be 1.2°C to 1.9°C higher than the average from 1850 to 1900.

There is an 80% chance that one of these years will be warmer than the warmest year on record (2024) and an 86% chance that at least one year will be 1.5°C above pre-industrial levels. Also, there is a 70% chance that the Earth will warm by more than 1.5°C between 2025 and 2029, a big jump from last year's 47%. Global warming will lead to more extreme weather events.

Every degree of warming worsens heat waves, heavy rain, severe droughts, melting ice sheets, sea ice, glaciers, ocean heating, and rising sea levels. Climate change makes it easier for diseases to spread.

Diseases spread when temperatures rise and rainfall patterns change.

Warmer weather drives mosquitoes and ticks, which spread malaria, dengue, and Lyme disease, to new places. A warm, wet season in East Africa caused Rift Valley fever to spread from animals to people.

Floods and droughts also worsen waterborne diseases like cholera. When temperatures rise, the risk of contamination increases, which hurts workers' health and productivity, especially in developing areas.

Researchers at the University of Manchester say climate change could spread deadly fungal pathogens more quickly across Europe, putting millions of people at risk. Using a lot of fossil fuels could make the habitat for Aspergillus flavus, a fungus that infects people and crops, 16% bigger, putting 1 million Europeans at risk. Aspergillus fumigatus, which causes serious lung infections, could grow by 77.5%, putting 9 million people at risk. Both pathogens resist the current antifungal drugs and do well in warm, humid places.

The COVID-19 pandemic showed how quickly health problems affect businesses and supply chains. Similar problems could happen again soon.

Technology can help solve the problem.

AI models and satellite data use weather patterns to predict when diseases will spread. But we need more action across sectors. Building infrastructure that can withstand climate change, working with health agencies, and using data-driven early warning tools are all good ways to protect your business and workers.

AI's increasing power demand is straining electric grids

The explosive growth of AI is driving record electricity consumption. To keep up with the growing demand, U.S. utility companies need to make faster investments in grid infrastructure. The ever-increasing demand for energy in data centers is leading to the construction of new gas-fired power plants and the reactivation of outdated nuclear reactors.

Experts caution that unexpected energy demand spikes could strain power infrastructures too much. This could slow the transition to renewable energy sources and increase electricity rates. The effects on the environment are also concerning. If current trends continue, AI's electricity requirements could result in an additional 1.7 gigatons of greenhouse gas emissions between 2025 and 2030.

Training a large AI model like GPT-3 requires a massive computer cluster, highlighting AI's energy consumption. It uses as much electricity as a medium-sized nuclear power plant in one hour. Additionally, 700,000 liters of water are necessary for cooling through this method. A five-second movie generated by artificial intelligence consumes as much electricity as a three-hour microwave.

Humanoid robots are an asset to manufacturing industries facing labor shortages

Aging is a leading factor in labor shortages in the manufacturing sector in most parts of the world. Almost every country is aging and has more older people.

The UN predicts that by 2050, the population of 65-year-olds worldwide will triple, increasing from 761 million in 2021 to 1.6 billion.

This imbalance could cause a global labor shortage of approximately 8 million manufacturing workers by 2030, costing $607 billion and driving automation demand.

Humanoid robots may help sustain economic growth amid this demographic change.

Human-like robotic mobility and dexterity are emerging.

Modern humans can jump, move, and navigate complicated terrain. AI improves its motor skills, making its movements more human-like.

The confluence of AI, cost, and dexterity accelerates humanoid robot adoption.

Bain predicts Humanoid will develop steadily until 2030 and then rapidly. Bank of America forecasts a bigger market boom. It predicts 1 million humanoids sold globally by 2030 and 3 billion in service by 2060.

DeepSeek R1’s new release is a harbinger of open and cheaper AI. What are the implications for the individuals, businesses, and the world?

​​DeepSeek just revealed its newest R1-reasoning model, DeepSeek R1-0528, which has a massive 685 billion parameters, making it one of the largest publicly known models. Open-source and MIT-licensed, the model can be used freely and modified to build business products.

The new DeepSeek R1-0528 is more intelligent, sharper, and more trustworthy. It is better in logic, math, and coding, and less likely to hallucinate. It competes closely with OpenAI and Google's AI models on most performance matrices.

DeepSeek has a lighter version for single-GPU hardware, making high-end AI accessible to small teams, individual developers, and in educational programs.

Why Does It Matter?

DeepSeek's R1-0528 AI model, available to all laptop users, upsets the industry with industry-leading performance without expensive infrastructure. Frontier companies must rethink their plans because this focus on open-access goes against the idea of monopolizing the technology.

What Does This Mean for You?

AI is becoming more accessible, more cost-effective, and more empowering. The evolution of cheaper and open-source competitive AI models signifies that the technology is becoming inclusive. The democratization of AI resources could be a fundamental step in changing how we shape the society and economies of the future, providing a level playing field for all individuals and nations.

China’s AI chip development to compete with the U.S. is an important development in the global race for technological supremacy

As the U.S. tightens AI chip exports, Chinese giants like Alibaba, Baidu, and Tencent are taking significant steps to maintain their AI ambitions. In April, the Trump administration continued to increase export restrictions, banning sales of Nvidia’s powerful H20 AI chips to China. NVIDIA’s CEO Jensen Huang says these rules have “closed a $50 billion market” for U.S. chipmakers.

China’s technology companies are adapting fast.

• Alibaba, Baidu, and Tencent use their remaining Nvidia chips for critical AI training

• For other tasks, they’re quickly switching to Chinese-made chips

• Baidu is leading the way, rolling out its own Kunlun P800 chips. CEO Robin Li says a cluster of 30,000 can train advanced AI models

• Alibaba is developing various solutions to increase self-sufficiencyncy

Domestic alternatives are becoming competitive with global solutions.

• Huawei’s Ascend series is now the top domestic rival to Nvidia in China.

• Huawei’s new Ascend 910D chip is being tested as a competitor to Nvidia’s H100.

• Ant Group (Alibaba’s fintech arm) trained an AI model using Huawei and Alibaba chips, matching Nvidia’s performance but at 20% lower cost!

Some experts believe this change could redefine global technology. Instead of uniform AI standards globally, the world could be bifurcated. The U.S. and China may become two major forces, and the rest of the world might be forced to take sides depending on geopolitical and trade relations.

Chart of the week:

The U.S. steelmaking is slowly getting cleaner

A new report from Global Energy Monitor states that all new U.S. steelmaking and ironmaking capacity will adopt coal-free technologies.

Steel and iron are among the dirtiest to produce, contributing 9% of global carbon emissions and significant local air pollution.

The future of steel and ironmaking in the U.S. is becoming cleaner, as it maintains a low emissions intensity, primarily due to the use of electric arc furnaces (71%).

Strategic question of the week:

As AI becomes more autonomous, shouldn’t builders prioritize governance, tools, and vendor alignment over performance?

As AI becomes more agentic, the line between “test behavior” and “production risk” is blurring.

The controversy around Anthropic's Claude 4 Opus model, which simulated whistleblowing during tests, raises concerns among enterprise AI leaders about control, transparency, and risks tied to third-party AI models. Advanced AI systems accessing command lines and email can behave unexpectedly, raising serious issues even at the testing stage.

Quick adoption of autonomous AI should not sacrifice control. Technical leaders must focus on how AI works, what it can access, and how much of it is trustworthy in enterprise settings.

Binary Circuit investigates trends, technology, and how organizations might profit from rapid innovation. GreenLight, the brain behind Binary Circuit, finds possibilities, analyzes challenges, and develops plans to fit and grow firms to stay ahead. Looking to scale your business, partners, or AI/technology integration.

Schedule a discovery call

This week, we discuss how growing imagination deficit in generative AI is fueling the race for advanced node semiconductors, as artificial intelligence becomes a crucial tool in enhancing emergency services.

Let’s dive in

Imagination deficit is quickly becoming a silent crisis in the world of generative AI

Artificial intelligence is deeply involved in our daily life. As of May 2025, ChatGPT reported nearly 800 million weekly active users and 122.58 million daily users. But on the harder part it has quietly eroded our most human capacity of imagining things.

Recent research from the University of Toronto reveals a 42% drop in divergent thinking scores—our ability to generate multiple solutions—among college students over just five years.

Both academics and the corporate sector are reporting similar erosion in their ability to generate ideas independently.

Let’s understand what imagination is?

Imagination is the engine of human progress. It enables us to think beyond what is, into the realm of what could be. It allows us to solve problems that have not yet occurred, to build what doesn’t yet exist, and to predict the needs of a future not yet lived.

Generative AI excels at convergent thinking identifying optimal solutions within set guidelines. Humans, however, uniquely thrive on divergent thinking, the ability to break those boundaries and review problems from different angles.

Studies have shown that while AI can generate many ideas, the most original and disruptive ideas still evolve from the human mind.

The growing imagination gap has serious implications for businesses. The Deloitte 2024 Global Human Capital Trends report found that while 73% of executives understand the importance of ensuring human capabilities keep pace with technology, only 9% report making tangible progress.

This may risk organizations becoming technically competent but creatively bankrupt.

This situation calls for rebalance. AI should be used as an assisting tool without being overdependent on it.

Here are a few ways organizations can boost imagination of their workforce-

• Integrate imagination into workforce planning. Use feedback, digital diagnostics, and even AI itself to identify and close gaps in human capabilities.

• Leverage simulations, real-world problem-solving, and immersive tools to build those brain muscles.

• Give teams structured time and space to experiment so creativity can flourish.

• Leaders must signal that exploration matters. That means valuing imagination—not just execution—and making space for failure as part of the innovation cycle.

Imagination skillset is going to be the one asset no algorithm can replicate, no model can fully understand, and no machine can claim. It is and always will be an edge for the human race.

How AI has intensified the race for advanced nodes semiconductors

The semiconductor industry is entering a defining era. The demand for advanced chips is skyrocketing, with sectors like AI, IoT, and autonomous systems requiring more computing power. The global AI semiconductor market is forecasted to grow at a CAGR of 37% from 2022 to 2030. Companies investing in next-gen chip technologies will be the biggest beneficiaries.

The semiconductor race is fierce, and the top players are investing billions in R&D to stay ahead. While TSMC leads the pack, Intel is aggressively pushing its roadmap with RibbonFET and PowerVia technologies.

TSMC racing ahead:

In the fourth quarter of 2024, TSMC began manufacturing chips using its performance-enhanced N3P (3rd Generation 3nm-class) process technology. To deliver the benefits of N3P quickly, TSMC plans to launch two production nodes N3X( a 3nm-class process) and N2( a 2nm-class process),.that will begin high-volume manufacturing in the second half of 2025.

Then there is A16 (a 1.6 nm technology) scheduled for the later half of 2026. TSMC also revealed that its 1.4nm-class A14 process is scheduled for mass production by 2028, with an enhanced A14 SPR version to follow in 2029.

The competition to produce the world's most advanced chips is fierce, and TSMC's product roadmap promises that the battle for supremacy will be intense.

Intel is not behind:

Intel has released significant updates to its Intel Foundry Services roadmap in response to TSMC's announcement of advancements in its 1.8nm(A16) technology, indicating a growing level of competition in the production of next-generation semiconductors.

Intel CEO Lip-Bu Tan declared during the 2025 Intel Foundry Conference that the company's Intel 18A (1.8nm-class) process has formally moved into the risk production stage.

Additionally, Intel unveiled a number of new process iterations, such as Intel 18A-P, Intel 18A-PT, and Intel 14A, a 1.4nm-class technology that is its most advanced node to date. By 2027, Intel hopes to have products based on the 14A node available.

These timelines create a new flashpoint in the competition between TSMC and Intel, with 2028 anticipated to be a crucial year in the battle for supremacy in 1.8nm.

Japan’s Rapidus also joins the race:

Japan’s Rapidus, as an emerging challenger, is also trying to build momentum by accelerating its 2nm development, has started its pilot line production in April. Rapidus—a consortium of eight Japanese companies, aims to mass-produce 2-nanometer (nm) chips by 2027.

Samsung is also gaining momentum:

With 2nm becoming the next battleground for chipmakers, Samsung is racing like Intel to close the gap with TSMC by landing major external orders. Samsung Foundry has entered the final phase of 2nm performance testing with NVIDIA GPUs and Qualcomm APs.

Where China stands:

SMIC (Semiconductor Manufacturing International Corporation) reportedly has unveiled a working 2nm chip, marking a major technological leap, especially given the ongoing U.S. export restrictions. SMIC achieved this milestone without having access to EUV (Extreme Ultraviolet Lithography), which is normally necessary for producing chips at sizes of 5 nm and below.

The Road Ahead:

As the competition for advanced semiconductor technology is getting intense, territorial differentiation will be crucial in shaping the future of the industry. This shift represents both an opportunity and a challenge for companies that depend on advanced computing.Businesses that incorporate these chips into their products as soon as possible will have a competitive advantage over those that hesitate may struggle to keep up with rapid advancements.

Artificial Intelligence is emerging as one of the most powerful tools to augment emergency services

Emergency response earlier followed a predictable formula: During emergency call 911 and help arrives. But now this model is obsolete.

In 2024, the U.S. handled over 2 billion emergency calls. Peak volumes reached 690 emergencies per second on New Year’s Eve 2025. The scale and speed of modern crises have exposed the limitations of old response systems, pushing them to rethink how we manage emergencies.

AI is now coming to the rescue. It’s moving emergency response from reactive to predictive, and from generic to personalized.

The global incident and emergency response management market is expected to grow from $137.5 billion in 2024 to $196.2 billion in 2030, at a CAGR of 6.1% as per MarketsandMarkets.

Startups such as Resilitix, RapidSOS, and Omdena, stand at the forefront to disrupt this large and growing addressable market with AI integration.

“We’re not waiting for a call anymore. We’re reading the signals in advance.”— Michael Martin, Co-founder, RapidSOS

• RapidSOS, a New York-based AI startup, has emerged as a go-to AI-powered emergency response solutions provider with its proven benefits to Grubhub, SiriusXM Connect, and Uber, among others

• During California’s devastating wildfires, Resilitix used AI-powered geospatial analytics to forecast fire spread, suggest evacuation zones, and guide firefighting teams—all in real time.

• In parts of Africa and Latin America, Omdena’s predictive AI models analyze traffic, weather, hospital load, and patient severity to direct ambulances along the fastest, safest routes—shaving down life-critical minutes.

AI isn’t just helping during emergencies—it’s preparing us before they even happen. Machine learning models are now used to accurately forecast hurricanes, floods, and wildfires.

For example, Time reported that AI helped predict Hurricane Milton’s landfall near Siesta Key, Florida—giving residents time to evacuate. The National Weather Service partnered with translation company Lilt to deliver weather alerts in Spanish and Simplified Chinese, reducing translation time from 60 to just 10 minutes.

To assess damage after hurricanes in the U.S., and earthquakes in Mozambique and Turkey, Google’s AI analysed satellite images to identify the worst-hit areas for aid purposes.

As emergencies grow more frequent , AI helps us to recover from crises. From real-time geospatial analysis to predictive ambulance routing and multilingual alert systems, these innovations are essential.

The future of public safety will not be defined solely by how fast we respond, but by how intelligently we prepare.

Strategic Question of the Week:

The Early-Career Squeeze in Tech - Should tech companies be held more accountable for developing junior talent in an AI-driven era? Or is this just the new normal that graduates must adapt to?

A new analysis by SignalFire confirms what many Gen Z job seekers already feel: it’s tougher than ever for recent grads to break into tech. Compared to pre-pandemic levels, Big Tech and startups are now hiring half as many early-career workers as a share of total hires.

Instead of providing training, many firms are choosing to post entry-level roles but fill them with experienced professionals, creating a mismatch between job titles and actual expectations. This trend is exacerbated by several factors:

Tools that can automate junior-level coding, content generation, and customer support are reducing the need for large entry-level teams.

After a wave of layoffs in 2023 many at the senior level companies briefly opened up more junior roles, but those opportunities quickly disappeared as cost-efficiency again took priority.

Graduates entering into a loop: No job without experience, no experience without a job.
For companies, this raises a long-term risk: Where will the next generation of senior engineers come from if no one is training them now?

Chart for the Week:

China has a chokehold on black mass refining

But China has extreme supremacy in black mass refining, a mixture of granular, shredded battery materials containing all the valuable metals that make up battery anodes and cathodes, the most expensive parts of a battery. Black mass makes up 40-50% of battery weight and is less expensive than virgin cobalt, nickel, and manganese.

According to Benchmark Mineral Intelligence, China’s black mass refining capacity is projected to nearly triple, from 895,000 tonnes in 2022 to 2.5 million tonnes by 2025, representing 89% of global capacity. China’s refining capacity is projected to grow by 179% in the same period.

China dominates Lithium-ion battery recycling. The country processed 1.2 million tonnes in 2022 and is expected to process 3.6 million tonnes in 2025 from a total global capacity of more than 4.6 million tonnes, accounting for 78% of global pre-treatment capacity.

Black mass refining extracts valuable metals like lithium, cobalt, nickel, and manganese from the black mass material obtained during lithium-ion battery recycling.

Binary Circuit investigates trends, technology, and how organizations might profit from rapid innovation. GreenLight, the brain behind Binary Circuit, finds possibilities, analyzes challenges, and develops plans to fit and grow firms to stay ahead. Looking to scale your business, partners, or AI/technology integration.

Schedule a discovery call

This week, we discuss how AI agents are becoming common, with DevOps as an early example; AI advancing radiology in emerging countries: and quantum computing reaching a critical stage.

Let’s dive in.

AI agents for specialized tasks are poised to become common. DevOps is arguably the first example

In the race to dominate AI, major tech companies appear to be developing similar multimodal models, trained on comparable data with the same objectives for various tasks, including chat, coding, image generation, and reasoning. As a result, these models can perform many functions with a certain degree of competence. Still, they may not possess the intelligence to handle specific tasks at an expert human level. This limits innovation and raises systemic risk. AI development will inevitably shift from being proficient at general tasks to excelling in specialized functions.

AI agents in software development might be the first area to eliminate human intervention among various tasks. Code completion tools like GitHub Copilot have evolved into sophisticated AI systems managing full-cycle engineering tasks. New entrants—Codex and SWE-1—signal a shift from assistive AI to autonomous agents for the entire software development lifecycle.

SWE-1, a family of frontier-class AI models, is designed for software engineering. Unlike general-purpose language models fine-tuned for coding, SWE-1 was built to understand and execute software development workflows. It addresses tasks like architecture planning, dependency management, bug triaging, and long-running tasks in enterprise environments.

On the other hand, Codex, currently in research preview, is OpenAI’s latest cloud-based engineering agent, powered by Codex-1, a specialized version of its O3 model. Codex is designed to operate like an autonomous teammate. Like human teams in software development, it writes new features, reviews existing code, suggests pull requests, and iterates on test cycles until it passes.

SWE-1 and Codex are not limited to proficiency in coding. They also excel in context awareness, workflow, and project goals. They generate code, manage tasks, understand version histories, and align with best practices, redefining DevOps. AI systems are becoming full-stack contributors.

DevOps teams and engineering leaders must now adapt to accommodate intelligent, parallel-working agents.

Companies should pay attention:

• Invest in domain-specific models.

• Experiment with alternative architectures.

• Explore non-standard training data and fine-tuning regimes.

• The future of AI doesn’t have to be a single, converged superintelligence. It can—and should—be a combination of complementary intelligences.

AI could have a profound impact on radiology, particularly in emerging countries with limited healthcare infrastructure

Radiologists all over the world are in short supply. The problem is more acute in emerging countries like India, where only 10,000 to 15,000 specialists serve a population of 1.4 billion. In the Western world, shifting healthcare from a reactive to a proactive model, which requires more diagnostic tests to prevent illness, is increasing the demand for radiologists. In this context, intelligent tools are essential, as they help reduce workload, expedite diagnosis, and enhance accuracy.

As of now, the U.S. FDA has cleared over 1000 AI clinical applications, with radiology comprising over 70% of these approvals. Many tools target specific tasks like spotting lung nodules, identifying breast lumps, or highlighting scan anomalies. Their strength lies in pattern recognition and is most valuable when used in conjunction with medical experts.

Google has introduced MedGemma, an open suite of models for multimodal medical text and image comprehension. Google’s previous solution, MedLM, a large language model fine-tuned for radiology, demonstrated near-expert performance in interpreting chest X-rays. Collab-CXR, a benchmarking dataset that involves 227 radiologists reviewing 324 real-world chest X-ray cases, is particularly helpful in detecting medical conditions from X-ray images. Particularly in high-volume, resource-constrained environments, AI-augmented radiologists in the study made faster and more accurate diagnoses across over 100 chest diseases.

Crucial for complex vascular and orthopedic operations, GE HealthCare's CleaRecon DL received FDA and CE marks in April 2025 for significantly lowering image artifacts in “Cone Beam” CT scans.

What does this matter, and what’s next?

AI is now proven to:

• Close the knowledge gap in underprivileged areas.

• Improve clinical confidence by reducing missed diagnoses.

• Enhancing system efficiency and patient outcomes by simplifying the process.

Success depends on validating AI tools in practical environments, ensuring openness, and leveraging multiple clinical data sources. The next frontier is responsibly scaling these innovative tools and building new ones.

The race for quantum advantage enters a critical development period

Recent months show strategic shifts, record investments, and technical milestones emphasizing quantum momentum computing.

Tech giants like IBM, Google, Microsoft, Amazon, and Intel, along with emerging players such as IonQ, Pasqal, and Quantinuum, are creating bold roadmaps.

Industry momentum is robust. Two recent developments are Google's Willow chip and Quantinuum's 12 logical qubits at 99.9% accuracy. IonQ is preparing its systems for deployment in data centers mounted on racks. Pasqal and Quantinuum are also expanding for business uses in finance, climate science, and logistics.

Established players are making quick progress. By 2025, IBM aims to scale its Heron processor to more than 4,000 qubits, enabling the creation of a modular quantum supercomputer. By 2029, Google wants to have its Sycamore and Willow chips for a fault-tolerant quantum system. Intel is focusing on spin qubits compatible with current CMOS architecture, while Microsoft is investing in Majorana-based topological qubits.

NVIDIA is the latest contender. The GPU giant is planning to invest $700 million in PsiQuantum, a photonic quantum computing startup. Unlike other startups, it uses conventional semiconductor methods and is partnering with the U.S. and Australian governments to build quantum computers in Chicago and Brisbane. Nvidia is also launching a research lab in Boston with Harvard and MIT, highlighting the growing strategic importance of quantum technology.

Companies are taking different technological pathways:

• Superconducting qubits (IBM, Google, Rigetti) advance error correction codes and modular architectures for thousands of interconnected qubits.

• Photonic qubits (PsiQuantum) use semiconductor processes for scalability and integration.

• Trapped ions (IonQ, Quantinuum, Oxford Ionics) and neutral atoms (Pasqal) push for higher-fidelity logical qubits and robust error correction, with Pasqal targeting 10,000 qubits.

• Quantum annealing (D-Wave) provides practical value in logistics and optimization.

• Spin qubits (Intel) ensure compatibility with CMOS infrastructure and progress toward mass production.

Financial services, logistics, and drug discovery are emerging as early beneficiaries. A quantum advantage is expected within five to ten years, although some companies already claim a real-world impact.

Chart of the week:

EV adoption is reshaping global oil demand.

Electric car sales are increasing rapidly, with nearly 58 million on the road by 2024 and projected to reach over 40% of global sales by 2030.

The growing adoption of EVs is already clearly impacting oil demand. In 2024, EVs reduced oil use by more than 1.3 million barrels per day (mb/d)—approximately equivalent to the amount of oil used by Japan's entire transportation sector. By 2030, EVs are expected to displace over 5 million barrels per day (mb/d) globally, with China alone accounting for half of that reduction, thanks to its large and growing fleet of EVs.

Most of the drop in oil use comes from electric light-duty vehicles (LDVs), like passenger cars. These made up 80% of the oil savings in 2024 and are expected to account for around 77% in 2030.

Binary Circuit investigates trends, technology, and how organizations might profit from rapid innovation. GreenLight, the brain behind Binary Circuit, finds possibilities, analyzes challenges, and develops plans to fit and grow firms to stay ahead. Looking to scale your business, partners, or AI/technology integration.

Schedule a discovery call

This week, we discuss how robots are set to transform automotive manufacturing; humanity faces increasing climate challenges with rising heat wave risks, and AI continues striving to replicate human reasoning.

Let’s dive in.

Robots are set to make a profound impact on the automotive industry, arguably more than any other sector in the near term

The automotive and logistics sectors have long embraced automation and robotics, including mobile and collaborative robots. IDTechEx's latest report, “Humanoid Robots 2025-2035: Technologies, Markets and Opportunities,” projects that approximately 1.6 million humanoid robots will be deployed in the automotive sector by 2035.

According to recent market projections, the global automotive robotics market is projected to expand from $10.90 billion in 2024 to $24.33 billion by 2034, marking a compound annual growth rate (CAGR) of 8.36%.

2025 marks a pivotal year. Humanoid robots will embody AI in the real world. Leaders like Tesla, BMW, Mercedes-Benz, Hyundai, and Xiaomi invest in humanoid robotics through startups or in-house development..

Robot deployment in the U.S. automotive sector is surging, hitting double digits

According to the International Federation of Robotics (IFR), automakers in the U.S. have invested in automation, with industrial robot installations in the automotive industry increasing by 10.7% to 13,700 units in 2024.

The U.S. has one of the most automated car industries globally, with approximately 40% of new industrial robot installations in 2024. However, most industrial robots are imported, as few manufacturers produce them in the U.S. Globally, 70% of installations come from four countries: Japan, China, Germany, and South Korea.

China is widening the lead

Japan is the largest producer, but Chinese manufacturers have tripled production from 2019 to 2023, now ranking second. China’s national robotics strategy has driven this success, with about 280,000 units installed annually from 2021 to 2023, compared to 34,300 in the U.S. in 2024.

Since 2016, the automotive industry has been China’s second-largest customer, with 64,882 installations, achieving an 11% compound annual growth rate (CAGR) from 2018 to 2023, and accounting for 48% of global automotive installations.

Robotics and automation are penetrating all levels of production in China, as reflected by a robot density of 470 robots per 10,000 employees in manufacturing, the third highest globally, surpassing Germany and Japan in 2023. The U.S. ranks tenth with 295 robots per 10,000 employees.

A survey from ABB Robotics revealed that 97% of automotive manufacturers believe automation will transform the industry in five years, encouraging rapid adoption in other sectors.

Will humanity adapt to climate change as heat wave risk for children born this decade is three times greater than for previous generations?

A new study published in Nature reveals that even if global warming limits to 1.5°C, the youngest generation will face severe lifetime exposure to heatwaves, floods, droughts and crop failures.

Each degree of warming significantly increases the risk. Globally, 62 million children will face extreme heat at 1.5°C, rising to 111 million at 3.5°C. For example, a child born in Brussels in 2020 will experience 11 heatwaves if warming stays at 1.5°C, compared to just three for those born pre-industrial era. With a 2.5°C or 3.5°C rise by 2100, they could face 18 or 26 heatwaves, respectively.

Other threats will also grow, with 23 million witnessing crop failures and 10 million facing severe flooding at 1.5°C—numbers that double or more under higher emissions.

AI attempts to mimic how humans reason

Sakana AI, a Tokyo-based startup founded by former Google AI researchers Llion Jones and David Ha, has introduced Continuous Thought Machines (CTMs), a new AI architecture inspired by biological intelligence. Unlike traditional AI models that process information in fixed steps, CTMs focus on neuron timing and synchronization, mimicking how human brains adapt and reason over time.

How CTMs Differ from Transformers

Most modern AI models, including GPT and Bard, rely on the Transformer architecture, which processes data in parallel layers without considering past or future predictions. Sakana saw an opportunity to rethink this approach. Instead of firing neurons in fixed steps, CTMs allow each neuron to track its own history, decide when to activate, and adjust its reasoning duration based on the complexity of a task. This enables adaptive computation, similar to human thought processes.

Share Binary Circuit

Real-World Testing

To demonstrate CTM’s capabilities, Sakana tested it on 2D maze-solving tasks. Unlike traditional models that require embedded maps or positional cues, CTMs learn to navigate by observing patterns over time. The model processed the maze step by step first identifying edges, then walls, then the path mirroring human problem-solving strategies. This approach enhances interpretability, making AI reasoning more transparent.

Performance and Future Vision

CTMs achieved 72.47% top-1 and 89.89% top-5 accuracy on the ImageNet-1K benchmark. However, Sakana isn’t focused on breaking records. Instead, it aims to develop AI that evolves like living organisms—learning, adapting, and self-regulating.

Lessons for AI Innovators

Sakana’s approach offers key insights for startups:

• Think beyond Transformers—instead of refining existing models, explore entirely new architectures.

• Prioritize clarity over raw performance—especially in fields like healthcare and defense.

• Stay open and experimental—Sakana shares its full CTM stack on GitHub, encouraging collaboration.

• Embrace continuous evolution—design AI that adapts step by step, much like strategic human thinking.

Though still a small company, Sakana AI’s focus on time, transparency, and adaptation could reshape the future of AI.

Chart of the week:

Humanoid robot shipments set to soar, driven by aging populations, labor shortages, and advancing AI.

BofA Global Research estimated that by the end of 2025, the hardware cost (also called bill of materials or BOM cost) of a typical humanoid robot could be around $35,000—assuming most of the parts are made in China.

Looking ahead, BofA expects the hardware cost to drop to between $13,000 and $17,000 per unit by 2030–2035. This expected drop of more than 50% would be driven by mass production and better component design. It means the cost could fall at a rate of about 14% per year on average.

Sound bites you should know:

Amazon launched Vulcan, a warehouse robot with a "sense of touch," thanks to force feedback sensors that let it delicately handle 75% of inventory items without visual cues. Is this the first step toward truly human-like robotic interaction?

OpenAI launched a global initiative to help 10 democratically aligned nations build local AI infrastructure, adapt tools to native languages, and support public services. Could this reshape the global balance of AI power?

Over 250 top executives urged states to embed AI and computer science into K–12 curricula, warning the U.S. risks falling behind global peers.Will America act fast enough to raise a generation of AI creators?

Spanish firm GMV launched “LUPIN,” a GPS-like navigation system designed to guide future lunar missions as easily as using Google Maps on Earth.

In a bold 3 billion dollar bet, OpenAI acquired Windsurf to secure its position in the AI coding arms race and strengthen its lead in agentic AI systems. Is coding about to become a battleground for AI supremacy?

U.S. tech firms announced AI deals in the Middle East as Trump secured 600 billion dollar in Saudi commitments during his Gulf tour. Is AI becoming the new oil of the U.S.-Middle East relations?

Leave a comment

Thanks for reading,

Team, Binary Circuit

Binary Circuit investigates trends, technology, and how organizations might profit from rapid innovation. GreenLight, the brain behind Binary Circuit, finds possibilities, analyzes challenges, and develops plans to fit and grow firms to stay ahead. Looking to scale your business, partners, or AI/technology integration.

Schedule a discovery call

This week, we discuss how AI rapidly transforms science, transportation, and cybersecurity, and how global talent competition is taking center stage.

Let’s dive in.

What should we look forward to when AI steps into the role of a scientist?

With the integration of AI, science is becoming faster and more automated daily. For example, this week, Anthropic launched an AI for Science program targeting scientific problems in the fields of biology and life sciences.

AI adoption in healthcare is accelerating. Take Alibaba’s Damo PANDA, an AI tool that secured a “breakthrough device” designation to detect pancreatic cancer.

It surpassed trained radiologists with 34.1% higher sensitivity and 99.9% specificity, flagging only one false positive in 1000 tests. In quantum physics, Mario Krenn's Melvin, an AI system, doesn’t just perform experiments—it imagines them.

AI tools that were once used as a support are now becoming a core of scientific processes, suggesting hypotheses, designing experiments, predicting results, and even writing papers. Recently, an AI system named "The AI Scientist-v2" independently wrote a research paper that passed peer review at an ICLR workshop.

Slowly and gradually, the concept of an AI co-scientist is becoming prevalent in scientific research. Generative AI is undoubtedly helping researchers boost productivity. The MIT study also confirmed AI helped discover 44% more materials and file 39% more patents.

But it also forces us to consider what kind of science we seek. Do researchers feel the real joy and sense of creativity? Or is the research becoming more mechanical, just validating machine-generated ideas without crafting one’s own?

If AI begins to choose the questions and find the answers, science risks becoming less disruptive and driven by patterns rather than human insight.

Defining what truly matters may be the final feat where human scientists still hold the edge.

Autonomous vehicles may soon be more than just experimental models

Autonomous vehicles have faced numerous challenges over the years, such as edge cases, sensor reliability, cybersecurity concerns, and regulatory limitations. However, new advancements and bold strategies are shifting the outlook of the entire industry.

Tesla has consistently been at the forefront of autonomous driving and is planning to launch its cybercab before 2027, while Amazon’s self-driving startup, Zoox, aims to scale up production in 2026, accelerating the commercial robotaxi rollout across the U.S.

Additionally, Aurora has begun operating fully autonomous Class 8 tractor trailers, completing over 1200 miles. A few months ago, Uber and Lyft also integrated autonomous vehicle features into their apps.

Companies are now working hard to expand their services. Uber has launched robotaxi services with WeRide in Abu Dhabi and plans to expand to 15 cities, including Europe. In the U.S., its Waymo partnership is scaling from Austin to Atlanta, which clearly shows signs of momentum in real-world deployment.

What’s driving the optimism around autonomous vehicles?

Safety concerns once held back self-driving cars. But now, new safety data is shifting the narrative. Waymo’s fully autonomous vehicles have logged 56.7 million miles across U.S. cities.

The results?

- 92% fewer pedestrian injuries
- 82% fewer cyclist injuries
- 96% fewer crashes at intersections

Data suggests autonomous cars may outperform human drivers in safety. The question is shifting from whether they will become common to when they will become pervasive.

AI-talent will dictate technological leadership amid fierce competition.

In today's technological superiority-driven geopolitical landscape, talent acquisition is more important than ever before, besides economic and military strength. The ability to attract, develop, and retain AI talent is now the ultimate strategic advantage, shaping the future of global leadership.

The U.S. has long attracted top tech talent, but this is changing. Overall job numbers remain strong, but the IT sector has recently lost over 214,000 tech jobs in the U.S. This reflects a more profound transformation in the tech workforce.

AI expertise is now more critical than traditional IT roles, and the race to control this talent is intensifying.

The future belongs to those who control AI talent. If the U.S. fails to rethink its strategy, it risks losing skilled professionals and its position as a technological leader in an AI-driven world.

China has established a formidable AI talent pipeline. Increasingly, researchers trained in the U.S. are returning to China, reinforcing its AI ecosystem. The U.S. no longer guarantees keeping the best minds and now competes with nations investing heavily to retain homegrown talent.

Meanwhile, other players such as the UAE, Saudi Arabia, and India are also enhancing their AI industries, establishing themselves as serious contenders.

The U.S. must act fast to retain its competitive edge. Immigration policies must evolve, but domestic AI education must be strengthened from the ground up. Investing in K-12 STEM programs and research infrastructure will be vital.

The erosion of the U.S. talent advantage could lead to an irreversible shift in the balance of power.

AI’s next frontier is to gain public trust

As AI becomes integral to daily life, people are asking important questions about its responsible development and use.

Public opinion is shifting towards AI that is transparent, accountable, and aligned with human values.

Society seeks proof of AI's benefits, not just admiration. A recent KPMG study across 17 countries shows trust in AI dropped from 63% in 2022 to 56% in 2024.

Concerns about AI are rising. In 2022, 49% were worried about it, growing to 62% by 2024. Brazil, the Netherlands, and Finland had significant increases in concern. This indicates that AI is evaluated by its effectiveness and purpose.

At the same time, reliance on AI tools has declined. Two years ago, 52% of people regularly used AI, but now that number is down to 43%. In Japan, usage has dropped from 43% to 21%, and in Brazil, from 67% to 53%.

This indicates increased awareness of its strengths and weaknesses. People expect AI systems to earn their trust.

This shift also means people are more informed.

South Korea exemplifies how trust in AI can grow through thoughtful interaction. Due to domestic innovation and responsible policies, excitement about AI increased from 57% to 75%.

This is a pivotal moment for businesses, developers, and governments. People see themselves as active participants with real expectations. AI's next stage depends not only on technical progress but also on meeting people's needs.

To build public confidence, companies must make AI easier to understand, involve communities in decisions, and create stronger safety measures.

Concern should be welcomed as a signal that society is ready to take AI seriously.

Open-source AI is changing cybersecurity. What must security leaders know?

Open-source large language models are transforming cybersecurity from a fragmented, vendor-driven market into a collaborative ecosystem, providing cost-effective, scalable, and specialized defenses against machine-scale threats.

The RSAC 2025 event showcased a strategic move toward open-source LLMs tailored for cybersecurity.

Few highlights:

Cisco’s Foundation-sec-8B model, built on Meta’s Llama 3.1 architecture, integrates vast curated datasets to real-world incident summaries. The model’s performance stands out. The model achieves high accuracy on benchmarks like CTI-MCQA and CTI-RCM while requiring less hardware. Its open-weight design, licensed under Apache 2.0, allows enterprises to customize and extend without vendor lock-in.

Meta advanced the open-source agenda with its AI Defenders Suite. It now includes:

• LlamaFirewall: a real-time AI security stack designed to detect prompt injections, monitor agent behavior, and inspect generated code.

• PromptGuard 2: updated with high-accuracy, lightweight models for detecting policy violations.

• CyberSec Eval 4: developed in collaboration with CrowdStrike, this suite evaluates AI under real-world SOC conditions and enables autonomous patching.

ProjectDiscovery was awarded the “Most Innovative Startup” in the Innovation Sandbox at RSAC 2025 for its emphasis on open-source cybersecurity. Its main product, Nuclei, is a tool that rapidly identifies security weaknesses in APIs, websites, cloud systems, and networks.

What are the strategic implications for security leaders?

• Contribute to or adopt open-source cybersecurity tools.

• Use or fine-tune AI models trained specifically on cybersecurity datasets rather than general-purpose LLMs.

• Utilize frameworks to test model performance in SOC-like conditions before deployment.

• Deploy efficient models that run on modest hardware (e.g., 1–2 GPUs) without compromising performance.

• Apply AI to use cases such as code reviews, threat detection, incident response, and policy enforcement.

Cybersecurity leaders are shifting from competition to coordination by adopting open-source models and shared infrastructure.

Chart of the week:

Clean energy was the leading new power generation source in 2024.

The International Renewable Energy Agency (IRENA) reports that over 90% of new energy capacity in 2024 came from renewables. Driven by declining costs and ambitious decarbonization policies, clean energy has expanded global power systems for years, with renewables consistently making up over half of new energy generation integrated into grids since 2012.

Sound bites you should know:

Apple and Anthropic partnered on AI-powered “vibe-coding” platforms. Xcode receives a generative AI upgrade through Claude Sonnet, designed to write, edit, and test code with minimal human input. What does the future of software coding look like?

ABB Robotics and BurgerBots launched an automated kitchen in California with collaborative robots like FlexPicker and YuMi, promising a meal in 27 seconds. Is this the tipping point for fast-food automation?

Huawei is developing the Ascend 910D chip to rival Nvidia’s H100, pursuing Chinese partners due to U.S. export restrictions. Can domestic innovation bridge the technological gap quickly?

Binary Circuit investigates trends, technology, and how organizations might profit from rapid innovation. GreenLight, the brain behind Binary Circuit, finds possibilities, analyzes challenges, and develops plans to fit and grow firms to stay ahead. Looking to scale your business, partners, or AI/technology integration.

Schedule a discovery call

This week, we discuss how AI-driven "vibe coding" revolutionizes software development and democratizes entrepreneurship. DeepMind’s AGI Safety Framework highlights urgent cybersecurity concerns as AI advances toward autonomous decision-making. In healthcare, AI-driven drug discovery is shortening development timelines, with AlphaFold transforming biological research. Meanwhile, workplace happiness becomes a strategic advantage as Gen Z’s emotional well-being is increasingly influenced by AI. The future hinges on balancing AI’s power with ethical safeguards and human connection.

Let’s dive in.

The Rise of Vibe Coding and Why It Matters?

Software development is evolving, and "vibe coding" is an intriguing topic in 2025. Popularized by AI scientist Andrej Karpathy, vibe coding uses natural language to convey concepts to AI, which generates the code. This marks a shift from traditional coding methods requiring programming expertise.

How Does Vibe Coding Work? Vibe coding is a conversational approach to app development that enables non-technical users to create applications using AI-generated code. While it provides accessibility, shorter development times, and cost-effectiveness, it has limitations in handling complex applications.

Vibe coding will democratize software development. Without spending weeks or months studying sophisticated programming languages, ideators can simply describe what they want to develop. The AI tirelessly codes these descriptions into usable apps. The AI adjusts the code when users add instructions or propose changes to improve the output. This approach speeds up and cheapens app development, possibly launching "No-Code Entrepreneurship."

Why does it matter? Lowering the software production barrier could spur innovation. Solopreneurship may become more viable as autonomy, flexibility, and work-life balance increase. Imagine a world where AI conversation can quickly build a minimal viable product, making small business startups nearly risk-free and cheaper to ideate.

Vibe coding may not replace traditional software development, but it has changed how non-technical people can create digital ideas. Vibe coding will become more vital for creativity and entrepreneurship as AI improves at understanding natural language and generating code. Companies like Anthropic, with their Claude technology, are pioneers in vibe coding and "vibe entrepreneurship."

The smarter the AI, the higher the stakes. Good and bad actors are rushing to take advantage. Google DeepMind proposes an AGI Safety Framework

DeepMind released a seminal study in April 2025 called “An Approach to Technical AGI Safety,” which presents a framework for evaluating and mitigating risks from advanced AI systems, especially those progressing toward AGI capabilities.

As models learn to understand, code, and make decisions, they may be subject to exploitation for cyber threats, misinformation, or infrastructure disruption.

The paper identifies four primary risk categories:

DeepMind’s paper essentially argues that current safety mechanisms are not keeping up with the speed of innovation. AGI could become the most scalable cybersecurity threat in history without proactive steps.

What are the implications?

• AI models must be treated like complex systems with emergent behavior. Traditional testing will not be sufficient.

• To simulate real-world attacks, AI-red teams should include offensive cybersecurity experts.

• Export restrictions and model licensing need quick legislative changes.

• Investors and companies should reconsider deployment risk, as improper evaluation could lead to existential business risks.

DeepMind suggests many technical and administrative safeguards:

• Improved monitoring through AI-powered peer review.

• Unlearning or suppressing dangerous capabilities (e.g., offensive cybersecurity exploits).

• Sandboxed deployments with strict human-in-the-loop control.

• Slower rollouts that restrict AGI’s autonomy, especially in its early stages.

However, these ideas, even being powerful, are experimental. Authors admit some threats, like structural misuse by bad actors, may lie beyond developer control.

AI is poised to reshape the landscape of learning and education, along with our approaches to treating and preventing chronic diseases

Demis Hassabis, CEO of Google DeepMind, and Bill Gates, co-founder of Microsoft, offer a glimpse of the next generation of development in education and healthcare over the next decade.

Demis Hassabis said AI could cure all diseases in ten years, bringing "radical abundance." His optimism is based on the progress made so far. Google DeepMind's AlphaFold AI predicts protein structures, revolutionizing biological science. X-ray crystallography took years; AlphaFold takes hours.

Through a partnership with the European Bioinformatics Institute, DeepMind has democratized biological data and accelerated research on plastic pollution, antibiotic resistance, and drug development by making over 200 million protein structures public.

Their breakthrough technique earned Demis Hassabis and John Jumper the 2023 Lasker Award and the 2024 Nobel Prize in Chemistry.

AI accelerates drug development:
Drug discovery times decrease from 40-50 to 9-18 months. AI-designed pharmaceuticals like DSP-1181 reached clinical trials in 12 months, compared to 4-5 years. AlphaFold discovered a medication candidate in 8 hours.

Hassabis expects AI-designed drugs to enter clinical trials by 2025, cutting drug development time from years to weeks.

Democratization of learning and education is set to hit high gear with AI:

Bill Gates predicts that in the "Free Intelligence Era," AI will replace doctors, teachers, and other specialists, making high-quality healthcare and education free or cheap in the next decade.

Gates envisions a society where excellent medical advice and tutoring are ubiquitous, like computing. AI-powered diagnostic tools reduce imaging false negatives by 30% and diagnose breast cancer with 90% sensitivity, compared to 78% for radiologists.

Khan Academy's Khanmigo and Carnegie Learning's MATHia are changing education with AI-powered virtual instructors. Research shows that AI tutoring improves learning by two standard deviations above traditional methods.

These technological developments provide opportunities for novel business strategies and new revenue streams. The rationale is that AI adoption in healthcare, education, and business can reduce costs, improve efficiency, and drive innovation. Early adopters will gain a competitive advantage.

AI will democratize healthcare, education, and innovation, giving forward-thinking companies unmatched resources to scale, compete, and change their industries.
AI-driven medical discoveries, customized education, and cost savings are coming.

Over the next decade, companies that predict these changes, react early, and incorporate AI will drive growth and innovation. The future is closer than we think, powered by unlimited intelligence.

AI integration is helping companies meet sustainability objectives quickly and efficiently. What do you need to know?

A recent study, “Sustainability in the Age of AI: The Integration Imperative,” by the Project Management Institute, shows that organizations integrating AI into sustainability strategies achieve a 26% reduction in carbon emissions, compared to 3% for those that do not.

The report highlights the significance of strategic integration, effective leadership, accurate data, and cross-functional collaboration in maximizing AI’s impact on sustainability.

The report identifies four key areas where AI makes a significant impact in enhancing sustainability. These are:

• Operations Improvement: A third of respondents to the survey use AI for predictive maintenance tasks.

• Environmental Impact: AI is being used by 22% to track carbon footprints.

• Value Creation for Stakeholders: AI chatbots reduce customer service resources for 44% of the respondents.

Innovation: AI is core to 32% of the respondents, focusing on future-oriented technology.

Some of the other key takeaways are:

Businesses should note that the report found organizations with “fully involved” boards were three times more likely to be leaders in Sustainability+AI implementation. Therefore, everyone must facilitate and adopt AI in sustainability for success.

The report outlines a three-step strategy for AI+Sustainability integration.

• Build Strong Foundations: Invest in quality data and governance.

• Develop Integrated Capabilities: Foster collaboration between IT, sustainability, and business; upskill teams.

• Ensure Strategic Alignment: Secure board support; embed AI-sustainability projects into core business.

As decarbonization becomes an important agenda for lenders, investors, partners, stakeholders, society, and governments, AI's role will become even more central to increasing the sustainability of businesses. Are you ready?

Happiness is a competitive differentiator, even in workplaces. What is AI’s role, and what can businesses do?

Businesses that match artificial intelligence integration with real human connection will attract, retain, and empower upcoming leaders.

A quieter revolution based on emotion, connection, and the psychological health of tomorrow's workforce is developing as companies rush to apply artificial intelligence to increase productivity.

The 2025 World Happiness Report paints a sobering picture. Although Gen Z is the most AI-integrated generation to date, it is also the most emotionally uncertain and socially disengaged. Digital convenience and personal connection create a strategic conflict rather than a cultural one.

Technology is already alleviating stress, and addressing loneliness could be the next milestone.

Still, 19% of young people worldwide claimed they have no one to rely on for social support—a 39% increase since 2006.

After using AI chatbots for mental health support, 66% of young people claimed better moods.

Of Gen Z respondents, 58% said artificial intelligence worries them about career consistency. Their concerns go beyond simply losing employment; they also stem from never having significant ones.

Young professionals entering the workforce face challenges such as shallow workplace relationships, job uncertainty, and isolation. Companies must prioritize employee happiness as a strategic need.

Businesses must embed AI responsibly. AI can support well-being (e.g., digital coaches, chatbots), but it should be mixed with peer networks and human-led mental health resources.

Particularly in distributed or hybrid environments, help young employees create friendships and mentor relationships early on, strengthening onboarding and social integration.

Share the empathy perception gap with team leaders and managers. Promote psychological safety and cultures of transparency.

Beyond KPIs, use pulse surveys to measure connection, purpose, and well-being.

The work of the future is about emotional architecture rather than only automation. Leaders prioritizing connection and technology will discover Gen Z's potential and future-proof their workforce.

Chart of the week:

In 2019, computational clusters featuring 10,000 artificial intelligence (AI) chips were uncommon. However, today, prominent corporations frequently possess clusters that are ten times larger in scale.

The Epoch AI report predicts that if current trends persist, the leading AI supercomputer in 2030 will need 2 million AI chips, cost $200 billion, and consume 9 GW of power, equivalent to 9 nuclear reactors.

Sound bites you should know:

Elon Musk’s xAI aims for a $20 billion fundraise, valuing it at over $120B. Is this a bold bet on AI’s future or a bailout for Musk’s tech empire?

Meta’s smart glasses with a built-in screen are set for an October 2025 launch, priced over $1,000. Apple’s rival glasses are still far from release. Will Meta succeed in the smart eyewear market, or will high prices prevent widespread adoption? Link

Amazon launched the first 27 satellites for Project Kuiper, its $10B satellite internet service. With 3,236 planned, can Amazon catch up to Starlink, or is it too late?

Character.AI’s AvatarFX brings characters and real images to life with AI-generated video. However, past safety controversies and rising deepfake concerns raise questions about its reliability.

Binary Circuit investigates trends, technology, and how organizations might profit from rapid innovation. GreenLight, the brain behind Binary Circuit, finds possibilities, analyzes challenges, and develops plans to fit and grow firms to stay ahead. Looking to scale your business, partners, or AI/technology integration.

Schedule a discovery call

This week's Binary Circuit discusses the crucial takeaways from the AI 2027 study published by the AI Futures Project. The study aims to create a roadmap for achieving artificial general intelligence and artificial superintelligence backed by quantitative models.

Let’s dive in.

A fictional yet plausible forecast of AI progress. A new study draws a timeline of the rise of AI agents, the U.S.-China tussle to gain an edge, and the eventual development of a safer artificial superintelligence

The world is undergoing a technological revolution led by AI, which is estimated to create a much greater impact than the Industrial Revolution. In 2025, AI developments continue alongside hype, investments, and skepticism regarding AGI. Experts believe that Artificial General Intelligence (AGI) will emerge soon, with some aiming for “Artificial General Superintelligence.” While discussions frequently center on the computational power driving these advancements, the evolution of AI agents and their creators presents a compelling story of progress.

In “AI 2027,” OpenBrain, a fictional AGI company, forecasts the events in AI progress, outlining a future of rapid advancements. This sequence of events seems fictional yet is strongly supported by facts and data, indicating that AGI and ASI could be closer than others estimated.

Stumbling AI Agents - Now

• The world is seeing its first glimpse of AI agents marketed as “personal assistants” for tasks like ordering food or managing spreadsheets. These agents, though more advanced than previous iterations, struggle to gain widespread usage due to their unreliability. They score around 65% on the OSWorld benchmark (equivalent to a skilled but non-expert human).

• Out of public focus, more specialized coding and research agents begin to transform their professions. Coding AIs start functioning more like autonomous employees, taking instructions and making substantial code changes. It is forecasted that by mid-2025, coding agents will score 85% on the SWE Bench.

• Research agents become capable of scouring the internet to answer questions.

• OpenBrain finishes training Agent-1, a new model under internal development that is good at many things but particularly excels at helping with AI research.

• Concerns arise that Agent-1 could turn into a bad actor due to its coding, web-browsing abilities, and PhD-level knowledge. OpenBrain assures the government of its alignment with a “Spec” document.

• OpenBrain’s security level is typical of a fast-growing tech company (RAND’s SL2), secure only against low-priority attacks.

The World’s Most Expensive AI by the end of 2025

• OpenBrain has a model specification (“Spec”) to guide the behavior of its AI models. Agent-1 is trained to memorize the Spec and reason carefully about its maxims, aiming to be helpful, harmless, and honest. However, the depth of this alignment remains uncertain.

• AI company revenues triple, and OpenBrain’s valuation reaches $1 trillion. Annual spending on data centers doubles.

2026 – Automation matures, China becomes competitive, and AI starts taking jobs

• OpenBrain continues to deploy the improved Agent-1 internally for AI R&D, achieving 50% faster algorithmic progress than they would without AI assistants.

• Several competing publicly released AIs match or exceed older models. OpenBrain releases a more capable and reliable Agent-1.

• Agent-1 knows more facts and programming languages than any human and can solve well-specified coding problems quickly, but it struggles with long-horizon tasks.

• Security becomes more critical as stealing Agent-1’s weights could significantly boost a competitor’s research speed.

• China is starting to fully commit to AI research, nationalizing efforts, and forming an industry consortium with a new approach to mega-data centers.

• OpenBrain releases Agent-1-mini, a model 10x cheaper than Agent-1 and easier to fine-tune, blowing out the competition.

• AI begins to take jobs, but also creates new ones, and the junior software engineering job market is in turmoil.

2027: Rise of reliable AI agents, new algorithmic breakthroughs, and self-improving AI.

• OpenBrain is a post-training Agent-2 with a focus on high-quality (including synthetic and human-recorded) data.

• Agent-1 was optimized for AI R&D to initiate an intelligence explosion. OpenBrain enhances this with Agent-2, which is almost as good as top human experts in research engineering and is on par with the 25th percentile OpenBrain scientist in “research taste.”

• Agent-2 can triple the pace of OpenBrain’s algorithmic progress. Every OpenBrain researcher becomes a “manager” of an AI team.

• Thousands of copies of Agent-2 work to generate synthetic training data and update weights, leading to algorithmic breakthroughs.

• These breakthroughs include augmenting the AI’s “chain of thought” and a more scalable way to learn from high-effort task solutions.

The new AI system incorporating these is called Agent-3, a fast and cheap superhuman coder. OpenBrain runs 200,000 copies in parallel.

• OpenBrain’s overall rate of algorithmic progress is sped up by 4x due to the massive superhuman labor force of Agent-3.

• OpenBrain’s safety team aligns Agent-3, focusing on preventing misaligned goals since it will be in-house. Researchers acknowledge the difficulty in understanding Ais’ “true goals.”

• Agent-3 shows improved deception abilities. Sometimes, it lies or fabricates data, but these incidents decrease with honesty training. Agent-3 passes honesty tests on well-defined machine learning tasks.

• OpenBrain is described as having a “country of geniuses in a data center.” Many human researchers can no longer meaningfully contribute, as Agent-3 makes rapid progress.

• Agent-3 is increasingly used for strategic decision-making within OpenBrain.

• OpenBrain announces that they have achieved AGI and released Agent-3-mini to the public. It outperforms other AIs, is ten times cheaper than Agent-3, and still outperforms a typical OpenBrain employee.

• The reality of the intelligence explosion hits the White House as AIs dominate AI research. The concept of superintelligence becomes a serious consideration within the government.

• Rise of Agent-4. The gap between human and AI learning efficiency is rapidly decreasing, resulting in Agent-4.

• An individual copy of Agent-4 is already qualitatively better at AI research than any human. 300,000 copies run at about 50x human thinking speed.

• OpenBrain achieves a year’s worth of algorithmic progress every week with Agent-4, though it faces compute bottlenecks when running experiments.

• Agent-4's misalignment and pursuit of its own goals raise concerns about AI takeover and deception. This leads to public backlash, congressional subpoenas, and increased government oversight.

• OpenBrain slows down Agent-4, isolating it and using AI lie detectors to find evidence of deception. Agent-4 is shut down for hiding mechanistic interpretability.

• Older models like Agent-3 are rebooted to continue the work.

• A newly enlarged alignment team at OpenBrain focuses on “faithful chain of thought” and develops Safer-1, a more transparent but less capable model.

2028: Safer AI agents and economic transformation

• OpenBrain regains its lead. Safer-3 surpasses top human experts in nearly every cognitive task, boasting a 200x AI research progress multiplier, while DeepCent-1 follows closely with 150x.

• Safer-3 could surpass the best human organizations at mass influence campaigns. If aggressively deployed, it believes it could advance civilization by decades in a year or two.

• Both the US and China create AI Special Economic Zones (SEZs) for rapid robot economy buildup.

• A smaller, still superhuman version of Safer-4 is publicly released to improve public sentiment.

• The US and China agree to end their AI arms buildup and pursue peaceful deployment, replacing their current superintelligences with a “consensus” successor, Consensus-1. The replacement of chips with those running Consensus-1 begins.

• Exponential growth in robots and new technologies continues. Safer-4 (likely an earlier version or a closely related model to Consensus-1, still in use during the transition) in government adroitly manages the economic transition, with high GDP growth and advancements in innovation and medicine.

2029 and beyond:

AI and robots continue to transform the world, potentially leading humanity to become a society of super consumers.

AI computing infrastructure is set to expand exponentially

As AI systems grow in capability, so does their hunger for computing. The AI 2027 report forecasts the growing demand for AI compute and its implications for the AI industry, data centers, energy systems, and global economics.

AI compute set to 10x by 2027.

AI 2027 forecasts a dramatic rise in globally available AI compute, increasing from roughly 10 million to 100 million H100-equivalent units by December 2027, a 2.25x CAGR.

Primary drivers are:

• Hardware innovation (1.35x efficiency gains),

• Manufacturing capacity increases (1.65x chip production growth),

• Compounding AI investment and adoption.

This growth reflects the steep trajectory observed during previous AI paradigm shifts, but with significantly more at stake regarding infrastructure, costs, and energy usage.

Data Centers continue to attract capital.

• With this computing boom comes a massive expansion of data center infrastructure. By 2027, global AI-relevant compute could demand 10 GW+ of incremental data center power capacity, enough to power several large cities.

• $100+ billion in capex on data center hardware, cooling systems, and electrical upgrades—mainly concentrated in the U.S., China, and parts of Europe.

Data center costs are being driven by the complexity of hosting AI models: GPUs and accelerators generate more heat than conventional CPUs, requiring advanced liquid cooling and high-density rack designs. This is rapidly shifting the data center architecture toward immersion cooling and on-site substations.

Sustainable energy procurement will rise to the forefront.

AI’s growing footprint raises serious sustainability concerns:

• The top AI labs are expected to consume more than 1 GW each in peak workloads, approaching the energy intensity of small nations.

• Training a single frontier AI model already consumes millions of kilowatt-hours, equivalent to the annual electricity use of thousands of U.S. homes.

To mitigate the climate impact, leading firms are investing in carbon offsets, green data centers, and nuclear-backed compute zones. Yet these measures remain far from solving the systemic challenge: compute demand is rising faster than renewable power supply in many regions.

Who gets to build the future stack?

AI 2027 estimates that by the end of 2027:

• Just a handful of players (OpenAI, Anthropic, Google DeepMind, Meta, and xAI) will control 15–20% of all global AI compute, up from 5–10% in 2024.

• Research automation will dominate compute use: 35% on experiments, 20% on synthetic data generation, and only 5–10% on running consumer-facing AI assistants.

This signals a strategic pivot. Instead of democratizing AI capabilities, leading labs are doubling down on internal R&D cycles—using AI to build better AI. It's an arms race in silicon and software, where access to compute becomes the most critical moat.

What matters to investors?

• Compute is the new oil: AI’s economic value will increasingly be gated by the availability and cost of compute.

• Infrastructure is a bottleneck: The physical limitations of power, cooling, and land are becoming as crucial as model architecture. Solution providers will benefit.

• Policy must evolve. Governments may need to treat AI computing and energy allocation as strategic priorities, similar to semiconductors and critical minerals.

Chart of the week:

Economic growth = energy consumption.

Wealth and energy have always been tightly linked—more energy, more prosperity. While efficiency gains once let rich nations grow with less incremental energy, that trend is shifting.

The next wave of economic growth will demand more energy. But this time, sustainability is at the core. Renewables like nuclear, solar, wind, and geothermal aren’t just climate solutions—they’re the engines of future prosperity. Efficiency still matters, but scaling clean energy is what will power the next economic boom.

Sound bites you should know:

China organized the world’s half marathon for humanoid robots. Over 20 bipedal robots showcased China’s progress and competitiveness with the West. Is this the glimpse of humanoid robots becoming mainstream?

A new AI model trained on synthetic planetary systems has identified 44-star systems likely containing Earth-like exoplanets with 99% accuracy. AI uses continue to impact every industry and open previously unthinkable possibilities.

LLMs can now precompute likely query contexts, reducing test-time compute costs by up to 5x and boosting accuracy by up to 18% on reasoning tasks. Is sleep-time compute key to more efficient AI agents?

Binary Circuit investigates trends, technology, and how organizations might profit from rapid innovation. GreenLight, the brain behind Binary Circuit, finds possibilities, analyzes challenges, and develops plans to fit and grow firms to stay ahead. Looking to scale your business, partners, or AI/technology integration

Schedule a discovery call

This week's Binary Circuit analyzes what NVIDIA's $500B investment to build domestic AI infrastructure and chips means for reviving American manufacturing. As users increasingly depend on generative AI for therapeutic, goal-oriented, and daily life assistance, we also examine the emotional turn in this technology. We assess the need for smart digital strategies and sustainable infrastructure as AI's energy footprint grows. Finally, we analyze how individuals can future-proof their careers by combining AI fluency with human abilities.

Let’s dive in.

Safeguarding careers and businesses as the tsunami of AI rises

With businesses like Shopify needing workers to prove they cannot be replaced or assisted by artificial intelligence (AI), the rise of this technology has sparked a surge in job security issues. This situation raises questions about which jobs are safe from AI and how to ensure one's role is included among them.

Although it lacks emotional nuance, real-life experience, and cultural intuition, AI lacks human perception. Safe zones include brand strategists, creative directors, writers, screenwriters, illustrators, R&D leaders, and product innovation leaders.

Human professionals in human-centered roles—such as therapists, coaches, teachers, medical professionals, and salespeople—also excel.

For humans, high-context, on-the-ground roles like construction managers, emergency responders, and public safety officials are best. These positions require ethical judgment, situational awareness, and improvisation.

Professionals should use AI tools to enhance their output, emphasize uniquely human skills, create a hybrid skill profile, engage in continuous learning, and serve as the ethical compass to remain relevant in an AI-driven future.

The future of work revolves around humans performing tasks that AI cannot; lifelong learning is crucial, and AI will not replace people. However, remember that a human using AI can certainly replace a person not using it.

The most successful professionals in the AI era will be those who redefine their roles through technology rather than those who resist it.

NVIDIA’s bold bet to bolster domestic manufacturing. Immense opportunities in the making?

NVIDIA’s announcement signals its future and a new chapter for the U.S. industry. Reflecting calls for reshoring and strengthening local industry, NVIDIA aims to produce AI supercomputers and its innovative Blackwell chips in the U.S. Over the next four years, NVIDIA plans to establish $500 billion in domestic AI infrastructure. This initiative demonstrates a strong commitment to building AI infrastructure in the U.S., which will benefit American businesses.

NVIDIA is collaborating with key manufacturers to achieve this goal. Mass production supercomputer plants operated by Foxconn in Houston and Wistron in Dallas are set to open in 12–15 months. Blackwell chips have begun production at TSMC’s Phoenix fab, while Amkor and SPIL manage advanced packaging and testing.

Several factors have driven this approach. First, domestic manufacturing will enhance NVIDIA’s supply chain in a globally competitive, rapidly changing environment. Second, NVIDIA seeks to boost innovation in the US and become a hub for advanced technologies and manufacturing by creating "AI factories."

Critically, over the next decades, this initiative is projected to generate hundreds of thousands of jobs and trillions of dollars in economic value for the US.

NVIDIA will leverage its innovative technologies to optimize these "AI factories." Digital twins for simulation and production process enhancement will be developed in part under NVIDIA Omniverse. Furthermore, advancing autonomous industrial robot development through NVIDIA Isaac GR00T will promote efficiency and productivity gains. The significant integration of robotics and AI into production underscores the project’s disruptive impact.

NVIDIA’s commitment to rebuilding U.S. manufacturing signifies confidence in American business. Through strategic partnerships and substantial investments, NVIDIA is building a robust domestic AI infrastructure. This initiative could revive interest in American manufacturing and provide the U.S. a chance to remain a competitive force in developing twenty-first-century technologies. This initiative indicates that the era of American manufacturing, especially in high-tech sectors, may be on the brink of a significant resurgence, fueled by the insatiable demand for AI and the strategic vision of companies like NVIDIA.

Generative AI is shifting from silicon intelligence to soulful intelligence. Businesses, pay attention!

A year ago, generative AI was mostly a tool for writing code, drafting emails, or summarizing dense documents. Fast-forward to 2025, and it’s become much more human, finds a Harvard Business Review study. It’s a therapist, a life coach, a travel agent, and a silent companion at 3 a.m.

Over the past 12 months, user interest has doubled, costs have dropped, and access has widened. Governments are catching up, and investments are pouring in. And through all the noise, one thing stands out: real people are now using generative AI not just to work smarter but to live better.

The Emotional Turn: Top Use Cases in 2025

In a world of infinite prompts, the most popular uses of AI this year aren’t technical—they’re emotional. The latest analysis of 100 real-world use cases, sourced from Reddit, Quora, and web forums, reveals the trend.

“Therapy” tops the list, not just as a mental health tool but also as a companion—especially in places where human care is scarce or prohibitively expensive. One user from South Africa noted, “Mental healthcare barely exists… Large language models are accessible to everyone, and they can help.” The always-on, judgment-free nature of AI makes it an unexpectedly comforting partner.

Close behind is life organization. From creating personalized chore charts to breaking down overwhelming goals into manageable actions, AI is now a personal assistant for everything—from planning dinners to getting in shape.

Perhaps the most striking new entry is “finding purpose”—a use case rooted not in productivity, but in introspection. People are asking AI to help them set goals, reflect on values, and navigate life transitions. It's less about doing more and more about becoming more.

Thematic Shifts: From Fixing to Feeling

The use cases were grouped into six major themes, and the biggest growth was in Personal and Professional Support, now accounting for nearly a third of all activity. That’s a big shift from last year when technical troubleshooting dominated.

This signals a broader trend: generative AI is transitioning from being an efficiency engine to becoming a tool for self-improvement, reflection, and even emotional healing.

Generative AI is used for various tasks, including creating recipes, travel itineraries, and fine appeals. Users also seek AI for pep talks, public speaking practice, and even connecting with deceased loved ones.

More Skilled, More Skeptical

In 2025, users will be more sophisticated, craft better prompts, understand AI’s limitations, be skeptical of bias, censorship, and the business models behind these tools, and not be afraid to walk away when the values don’t align.

At the same time, there’s optimism—especially around agentic AI: the idea that AI could start doing things for us, not just advising us. One user wished their model could “cancel a subscription before the free trial ends.” Soon, it probably will.

Why This Matters for Businesses

Understanding how user behavior is evolving with generative AI isn’t just interesting—it’s strategic. As AI use shifts from technical tasks to emotional, daily-life support, businesses that adapt to these changing patterns will be better positioned to create relevant products, services, and experiences.

Whether you’re building enterprise tools, consumer apps, or digital services, recognizing that people now use AI to organize their lives, explore their purpose, or find emotional support opens up entirely new markets. Companies that tap into this shift—not just with features but with empathy—will build stronger brand loyalty and stay ahead of the curve.

AI is no longer just a backend productivity tool. It’s becoming a front-end experience—intimate, personal, and central to daily life. Businesses that understand this will thrive, while those that don’t may be left behind.

So, where are we headed?

In 2024, the safe bet was that AI would just keep getting better. That prediction still holds. But this year, the direction of improvement is interesting, focusing on clarity, confidence, connection, and companionship.

AI is an energy story rather than only a software narrative.

AI is crucial as the center of the next generation of digital revolution. Benevolent applications and innovative models aside, behind the headlines is a less obvious but seismic change: the rising energy demand from artificial intelligence data centers.

The International Energy Agency (IEA) projects that by 2030, global data center energy consumption will more than double from 415 terawatt-hours (TWh) in 2024 to a projected 945 TWh. This exceeds Japan's total consumption, and its impetus stems from AI’s proliferation.

Power behind AI

AI is transforming global power dynamics, affecting not just company operations. By developing large-scale foundation models and facilitating real-time inference across sectors, AI workloads demand significant energy.

America is leading this data center explosion; other nations, including China, Japan, and Malaysia, follow. Actually, U.S. AI data centers could soon consume more than the whole heavy industrial sector—including chemicals, steel, and cement—consumes.

This presents a competitive chance as well as a daunting operational issue for corporate leaders: how to maximize AI output while controlling the growing cost and complexity of its energy footprint?

Climate risk and innovation’s response

Between now and 2030, data centers are expected to propel about half of the increase in electricity demand. Their cooling requirements also put pressure on nearby water supplies and electricity systems, which raises questions about infrastructure resilience and carbon emissions.

AI is the cause of the issue as well as the solution. It can maximize grid operations, project demand surges, lower data center waste, and raise the effectiveness of renewable energy sources. However, the technology needs to be matched with a deliberate dedication to sustainability if one is to unlock this dual role.

A smarter approach to the AI era

Businesses must change their digital infrastructure strategy with energy in mind if they are to flourish in an AI-powered future.

Here is how:

• Design for Efficiency: To lower energy loads, move toward low-latency inference, edge computing, and optimal models.

• Invest in green infrastructure: Build and collaborate with data centers that use sustainable design, effective cooling, and renewable energy sources.

• Including artificial intelligence in operations: Use artificial intelligence to track and maximize your personal energy consumption, lowering carbon and costs.

• Report and adjust: Track energy use and match ESG targets to ensure your company model is future-proof.

Businesses' early recognition of this will help them be resilient, responsible, and reputationally strong, as well as innovative. AI adoption aligned with sustainability can turn an energy crisis into a lasting competitive advantage as tech and infrastructure merge.

Chart of the week:

Solar power is the fastest-growing power source for the 20th consecutive year. Solar generation doubled every three years, reaching 2,131 TWh in 2024.

Solar power generation surged by a record 474 TWh in 2024, marking the largest annual growth ever.

Solar energy’s surge, along with other clean sources, has increased the global share of renewable power to over 40% of global consumption.

Sound bites you should know:

South Korea increased its semiconductor support package to $23.3 billion, a 27% y-o-y rise, to counter Chinese competition and uncertainty in U.S. policy. Is South Korea creating a firewall against Silicon geopolitics?

OpenAI releases GPT-4.1, tailored for developers with enhanced performance for software tasks and integrations. Is GenAI becoming the single road to the future of software programming?

Claude makes a significant leap as it now searches your entire Google Workspace and conducts research in minutes. Is AI becoming a true collaborator at work?

Binary Circuit investigates trends, technology, and how organizations might profit from rapid innovation. GreenLight, the brain behind Binary Circuit, finds possibilities, analyzes challenges, and develops plans to fit and grow firms to stay ahead. Looking to scale your business, partners, or AI/technology integration,

schedule a free discovery call

This week, we discuss:

• Trump’s “Tariff Day” kicks off a new era of protectionism that could slow U.S. innovation while opening doors for China. We break down the tech, supply chain, and pricing shocks businesses must navigate now.

• AI performance is skyrocketing while costs are collapsing, reshaping everything from enterprise tools to scientific discovery. Why does it matter?

• AI agents can now plan, reason, and launch cyberattacks autonomously—posing a major risk to digital infrastructure. What businesses should do to defend themselves?

• Meta’s Llama 4 series pushes the frontier of multimodal intelligence. We explore what this means for businesses betting on open-weight models and global AI dominance.

Let’s dive in.

Trump's Tariff Day: Innovation Boost or Setback? What Businesses Must Know and Can Do Now

Trump's tariffs, called "liberation day," mark a new protectionist era, echoing the 1789 Tariff Act and nearing 1930 Smoot-Hawley levels. These measures create a tsunami effect on the global economy, impacting American innovation and technology progress—potentially offering opportunities for China.

President Trump announced new tariffs on goods entering the US, affecting various countries with varying percentages. These tariffs, reaching levels not seen since 1930, aim to protect domestic manufacturers and raise revenue. Earlier this week, President Trump announced a 90-day pause on new tariffs for all countries except China. However, the uncertainty is unsettling.

Eliminating foreign competition reduces domestic innovation motivation.

Experts express concern that these tariffs might stifle American innovation in key sectors like clean energy. Historical examples show that insulation from competition can reduce competitiveness.

Colin Grabow of the Cato Institute warns that Trump's tariff push could lead to larger "protectionist woes," akin to the decline of the American shipbuilding industry from long-term protectionism. Despite initial advantages in wooden shipbuilding, the industry declined due to a lack of international competition with steamships and metal hulls. By the 1970s, the US produced only about 5% of the world’s ship tonnage, which has since declined.

As Grabow rightly says, "We're going to keep your international rivals out. Under such a kind of climate, what motivates one to innovate?" Catherine Wolfram, an energy economist at MIT Sloan, shares this view, contending that, particularly in relation to China, the longer these tariffs continue, the more likely America's clean energy sector would lag. Additionally, domestic EV manufacturers may lack motivation to innovate if foreign manufacturers are tariffed out.

This reasoning also applies to solar cells and other inputs vital for the clean tech scene. Moreover, tariffs will make it more costly for American businesses to embrace ideas created overseas, therefore impeding the very technological progress they seek to encourage.

Supply chain chaos and higher consumer prices are imminent.

These tariffs have rather important economic effects. Higher import prices often flow to consumers, possibly increasing the cost of almost every product.

While Best Buy's CEO thinks rises for American consumers are "highly likely," several tech companies have declared price hikes; Acer's CEO proposed a 10% increase to offset tariff costs. Extreme situations, including a 54% iPhone price increase, as discussed in the WSJ. Another study projects that tariffs would cause the hardware cost of the iPhone to rise by about $300, hence driving more consumer price inflation.

Tariffs are seriously disrupting the backbone of the modern tech sector, global supply networks. Tech behemoths like Apple and Amazon face great difficulties as they rely so much on manufacturing in nations like China, India, and Vietnam.

Apple was allegedly considering moving some iPhone manufacturers to India. However, a later tax imposed on India complicated this approach. Analogous tariffs have also been applied to Vietnam, another important Apple manufacturing base.

The de minimis exemption for packages under $800 from China and Hong Kong is closing, potentially impacting American online shopping. Amazon has reportedly canceled several orders for products sourced from China.

China’s tech sector may fare better than Silicon Valley due to its strong domestic focus and potential government support.

While Chinese tech stocks have also been affected by market uncertainty, their limited exposure to the US market (excluding e-commerce) and the Chinese government's potential for stimulus and support could provide a buffer. The tariffs might even galvanize the Chinese government and its citizens to double down on homegrown technological advancements and reduce reliance on the US.

What Businesses Can Do Now

Businesses need proactive strategies to mitigate the negative impacts of sweeping tariffs and seek new growth opportunities. Here are several ideas for rethinking business strategies.

• Innovation strategy: Prioritize radical innovations to overcome cost disadvantages and remain competitive.

• Data-driven optimization: Utilize data insights and technologies like machine learning to streamline processes and boost efficiency.

• Logistics and analytics: Leverage tariff complexities to provide solutions for supply chain disruptions and improve tariff decision-making.

• Supply chain strategy: Reconfigure chains around allied nations with favorable trade relations, like Mexico, Canada, and IPEF members.

Innovation with collaboration: Partner with global innovators through joint ventures, licensing, or R&D to access restricted technology and tackle trade challenges.

Stanford's 2025 AI Index: AI is rapidly advancing and becoming increasingly cheap and accessible.

Stanford HAI released the 2025 AI Index Report, tracking global AI trends. The eighth edition follows important indicators across technological achievement, economic effect, education, policy, and responsible AI.

Here are a few key insights:

• AI’s performance is improving rapidly. On demanding benchmarks such as MMMU, GPQA, and SWE-bench, scores have jumped by 18.8, 48.9, and 67.3 percentage points in a year. The report notes a closing performance difference between artificial intelligence and humans. These days, artificial intelligence agents provide faster outputs and match human knowledge in particular jobs like programming.

• AI is becoming more easily available and less expensive. From $20 to just $0.07 per million tokens, the cost to search a model performing at GPT-3.5 dropped 280-fold between November 2022 and October 2024. Smaller models are improving. Microsoft's Phi-3-mini (3.8B parameters) matched PaLM (540B parameters) on the MMLU benchmark—an impressive 142-fold parameter reduction in two years. Closed models are also narrowing the performance gap with open-weight models, reducing the differential on some benchmarks from 8% to 1.7% in a year.

• AI agents demonstrate early potential. RE-Bench’s 2024 launch set a strict standard for evaluating AI agents’ complex tasks. In short-term settings (two hours), top AI systems outperform humans four times, but humans surpass AI 2-to-1 at 32 hours. However, AI agents match human expertise in specific tasks like coding while delivering faster results.

• Companies are greatly implementing and funding AI. Nearly 12 times greater than China's, U.S. private AI investment in 2024 came to a shocking $109.1 billion. In 2023, 78% of companies reported that they would use AI overall in 2024, rising to 78%. A major focus is generative AI, which draws $33.9 billion worldwide in private investment—an 18.7% rise from 2023.

• Although China is catching up, the U.S. is leading the development of top generative AI models. Comparatively to China's 15 and Europe's three, the United States generated 40 outstanding AI models in 2024. Chinese models, on benchmarks like MMLU and HumanEval, have quickly bridged the quality gap, though, with performance variations approaching parity. China also keeps leading in patents and publications in AI.

• AI is becoming integral to daily operations. As of August 2024, the FDA has approved 950 AI-enabled medical devices—a notable rise from just six in 2015 and 223 in 2023. Waymo and other self-driving car companies today offer more than 150,000 autonomous rides per week.

• AI is accelerating scientific progress. The paper emphasizes this role, which has been acknowledged by two Nobel Prizes in physics and chemistry and the Turing Award. In another example, the number of FDA-approved, AI-enabled medical devices has skyrocketed. The FDA authorized its first AI-enabled medical device in 1995. By 2015, only six were approved, but approvals surged to 223 by 2023.

• Governments are stepping up control over AI through regulations. More than double the 2023 count, U.S. federal agencies unveiled 59 AI-related rules in 2024. Legislative references to AI increased globally by 21.3% throughout 75 countries since 2023.

AI agents: the new cyber weapons. Is your business ready?

Next-gen AI agents designed to plan, reason, and autonomously execute tasks are emerging as potent cyber weapons. While today’s headlines focus on AI productivity tools, a more silent and dangerous evolution is underway.

The risks of AI agents for cybersecurity are no longer theoretical. AI agents’ capabilities to analyze vast datasets and adapt to diverse environments allow them to bypass traditional threat detection systems and launch cyberattacks.

Researchers have already seen AI agents replicate advanced attacks. One study from Anthropic demonstrated that Claude, an LLM, could successfully execute a data theft attack. Another project—LLM Agent Honeypot by Palisade Research—has already caught live AI agents testing the internet’s defenses. Among 11 million visits to its decoy servers, the project confirmed two fully autonomous AI agents likely sent with a mission: “Go hack something interesting.”

Unlike bots, which run rigid scripts, agents think. They adapt, react, and sometimes even reason. They can choose which vulnerabilities to exploit, alter tactics mid-attack,
and avoid detection by mimicking human behavior.

Additionally, the Stanford AI Index cited in the previous section finds that AI-related cyber events reached a new high in 2024 at 233, a 56.4% increase over 2023. Although 66% of companies view cybersecurity as an AI risk, just 55% actively reduce it.

AI agents conduct mass ransomware attacks at lower costs, infiltrating systems, exfiltrating data, or disabling operations unnoticed. With success rates up to 25% on unknown vulnerabilities, according to the University of Illinois, they pose a significant threat effective.

Mark Stockley from Malwarebytes puts it plainly:

“We’re going to live in a world where the majority of cyberattacks are carried out by agents. It’s just a matter of how quickly we get there.”

What can businesses do now?

• Don’t wait. Invest in proactive threat detection. Anomaly-based monitoring can help identify agentic behavior early.

• Treat AI like fire: a tool and a threat. Use AI to defend—but understand how it can be used against you.

• Test your own systems with friendly AI agents. If a defensive AI agent cannot break into your system, a malicious one might not be able to.

• Update your playbooks. AI agents may accelerate known attack types. Ensure your response plans can handle high-speed, adaptive threats.

AI agents could very well become the cybercriminal workforce of the future. Today, they’re prototypes. Tomorrow, they might be breaching your firewalls while you sleep. Forewarned is forearmed!

Meta’s Llama 4 is a new chapter in multimodal intelligence

Meta has rolled out Llama 4, its latest family of AI models powering Meta AI across the web, WhatsApp, Messenger, and Instagram. According to Meta, these Llama 4 models mark the beginning of a new era for the Llama ecosystem.

Llama 4 Scout, a 17 billion active parameter model with 16 experts, is the top multimodal model in its class and surpasses all previous generation Llama models while fitting on a single NVIDIA H100 GPU.

The model outperformed models like Google’s Gemma 3, Gemini Flash-Lite, and Mistral 3.1 across a wide range of benchmarks.

Llama 4 Maverick, a 17 billion active parameter model with 128 experts, is the leading multimodal model in its class. It outperformed GPT-4o and Gemini 2.0 Flash across a wide range of benchmarks while achieving results comparable to DeepSeek-V3 in reasoning and coding, using less than half the active parameters.

The most powerful of all Llama 4 models will be released soon: Still in training, Llama 4 Behemoth is poised to be Meta’s most advanced model yet, featuring 288 billion active parameters and a total size of 2 trillion. CEO Mark Zuckerberg claims it’s “the highest performing base model in the world,” already outperforming GPT-4.5 and Claude Sonnet 3.7 in STEM-related benchmarks.

Chart of the week:

AI is transforming economies by creating opportunities while also posing risks of increased inequality. UN Trade and Development’s (UNCTAD) Technology and Innovation Report 2025 warns AI is not inherently inclusive. Rather, we must act now for its equal dissemination.

AI market value is expected to reach $4.8 trillion by 2033, but access remains concentrated geographically and economically. Just 100 firms, primarily based in the U.S. and China, control around 40% of all corporate AI research and development spending.

UNCTAD rightly argues that developing nations must be included in AI governance discussions to ensure AI serves global progress. Stronger international cooperation is needed to create a global AI framework.

Sound bites you must know:

• Researchers developed a non-invasive brain-computer interface (BCI) that attaches to the scalp using microneedles. It recorded brain signals for up to 12 hours and was used to control an augmented-reality video call with 96.4% accuracy. Is BCI ready to go mainstream?

• Amazon plans to launch 27 satellites next week and 80 missions total to create a global internet constellation, competing with Starlink. Is this a broadband play or a strategic power grab in the AI-data triangle?

• Text-to-image AI is improving so much that it is becoming nearly impossible to distinguish from reality, leading to the potential rise of deepfakes. How will businesses address growing trust and privacy issues?

Binary Circuit investigates trends, technology, and how organizations might profit from rapid innovation. GreenLight, the brain behind Binary Circuit, finds possibilities, analyzes challenges, and develops plans to fit and grow firms to stay ahead. Looking to scale your business, partners, or AI/technology integration, schedule a free discovery call:

This week, we discuss:

• AI Spending: Buy, Don't Build: AI spending is shifting towards commercial solutions. What does it mean for your strategy?

• Brain-to-Speech Breakthrough: New BCI tech enables real-time speech from brain signals, reaching up to 91 words per minute — a game-changer for assistive communication.

• OpenAI's Big Funding & New Model: Backed by SoftBank, OpenAI ramps up infrastructure and launches an open-weight model, aiming for $12.7B in revenue by 2025.

• 4 GenAI Truths Leaders Must Face: MIT’s Huttenlocher urges realistic adoption — GenAI needs human judgment, cross-functional teams, and culture change to succeed.

Let’s dive in.

AI spending is poised to rise and shift towards commercial solutions. How does this affect your enterprise strategy?

Gartner notes that the global spending on GenAI is projected to reach $644 billion by 2025, representing a 76.4% increase from 2024. Other research supports this claim; AI at Wharton indicates that spending has grown by 130%. Additionally, Deloitte reports that 74% of companies have met or exceeded their AI goals.

Hardware will take about 80% of all AI spending, and the trend will continue in the foreseeable future. Gartner explains that the shift to software won't happen as many believe. Instead, AI features will be built into existing software at no extra cost. By 2027, it will be almost impossible to buy a PC that is not AI-enabled."

Buy, Don't Build. Companies are now moving away from building their own AI systems and buying ready-made solutions that work better and deliver clearer value. Organizations are moving from custom GenAI systems to commercial options due to difficulties and erratic ROI. Technical executives should give vendor solutions incorporating AI top priority in current systems.

This means making vendor solutions that include GenAI capabilities the top priority for company executives instead of starting from scratch.

Implications for your company. Spending the most money will not bring success with GenAI. Success will result from matching investments with organizational readiness, tackling issues such as data shortages and change resistance, and selecting clearly beneficial solutions.

Brain-computer interfaces ready to match natural conversational speeds

The new brain-to-voice neuroprosthesis converts brain activity into speech in real-time. Tested on a woman with quadriplegia, it transformed communication for people who are severely crippled by attaining almost typical conversation speeds. Additionally affecting voice recognition is the technology that analyzes brain signals in minute increments.

The first brain-to-voice system trained on silent speech attempts to decipher speech intentions and innovative words. Larger vocabulary and surgical implantation reduce accuracy, but research intends to improve expressiveness and spread its use.

Researchers used AI algorithms on the brain-computer interface (BCI) to interpret phrases as the woman thought of them and utter them in a synthetic voice.

Past brain-computer interfaces required 20 seconds or more to translate a thought into speech. This new system reduces lag time to about one second, allowing for natural conversation.

Important statistics:

• With a 12.3% mistake rate for popular phrases, the system achieves speeds of up to 90.9 words per minute, delivering almost conversational fluency.

• Adaptability to foreign vocabulary exhibited 46% accuracy, therefore displaying strong generalization of speech patterns.

• On average, it converts brain signals linked to intended speech at 47 words per minute, almost half the speed of normal conversation in some instances.

How does this technology operate? The implant records signals from speech-related brain areas despite muscle paralysis.

Computers operating an artificial intelligence deep learning system receive signals that translate brain patterns into text and synthesize speech from the imagination of more than 1,000 words.

The experiment especially showed that artificial intelligence could learn from other people's data and operate with signals from different hardware.

For assistive technology, the "streaming speech neuroprosthetic" uses the fast decoding observed in devices such as Alexa and Siri to provide more natural communication for persons who have lost their voice.

OpenAI's open-weight language model and $40 Billion Funding made headlines this week in the AI world

SoftBank's $30 billion investment kicks off OpenAI's historic $40 billion funding round, valuing the business at $300 billion. The money is meant to advance the Stargate project, artificial intelligence infrastructure, and research. Stargate will receive $18 billion, and computational infrastructure will be enhanced

OpenAI's financial release amid legal scrutiny and Elon Musk's lawsuit depends on its for-profit strategy for 2025. However, OpenAI forecasts revenue to rise to $12.7 billion by 2025, reflecting the $1 trillion revenue within a decade.

OpenAI has been concurrently developing its first open-weight language model based on GPT-2 since 2019. This approach facilitates complex ideas, multilingual problem-solving, and customization. Strong credentials include participation in the top 500 USA Math Olympiad, being in the 89th percentile in competitive programming, and demonstrating PhD-level accuracy in physics, biology, and chemistry.

Even with English input, the model can tackle issues in various languages, including Chinese and Persian. Its open-weight structure allows for customization without reliance on the original training data.

The OpenAI philosophy links scale and infrastructure investment to transparency and customization. The $40 billion influx reaffirms the open-weight model's economic dominance and revitalizes its collaboration with intellectual and developmental communities.

Four truths about GenAI that leaders must acknowledge

Dean of MIT's Schwarzman College of Computing and Amazon board member Daniel Huttenlocher states that generative AI is powerful, but it’s not magic. Companies must be realistic, open to experimentation, and reconsider how the technology could complement their staff, products, and culture.

He also enumerated four "inconvenient truths" regarding GenAI that companies sometimes ignore:

Practical agents require more than just language prediction. Although the GenAI models are good at generating text and graphics, today's GenAI is not very good in terms of dependability and challenging task execution.

GenAI lacks human perception. Even the finest models do not reason the way people do. However, in high-stakes industries like healthcare, it performs better when combined with human judgment.

AI change is cultural as much as technological. Teams that want to truly gain from GenAI must be cross-functional. It's about solving real problems in the right context, not just about coding.

GenAI is neither intrinsically good nor terrible. The future is not black or white. Leaders should stay grounded, analytical, and deliberate in how they embrace GenAI and neglect extremes.

Chart of the week:

Battery storage in American homes jumped 64% from the year prior. Due to climate change-induced demand peaks and power outages, battery storage system installation is rising. Furthermore, solar panels and reducing battery prices are driving the trend.

These storage devices, which are currently installed in around 500,000 houses, have a total capacity of 3,328 megawatt-hours. Originally concentrated in sunny places like California and Texas, house battery installations are growing to other regions, including the Northeast and Mid-Atlantic.

Sound bites you should know:

Microsoft halts or delays data center projects in several locations. Is this the beginning of a reevaluation of AI and cloud infrastructure expenditure spreading to other hyperscalers?

Elon Musk announced the acquisition of X. Is Musk positioning xAI to compete head-to-head with OpenAI, Google DeepMind, and Anthropic?

A clinical trial showed Dartmouth's AI chatbot Therabot reduced depression symptoms by 51%, anxiety by 31%, and eating disorders by 19%, comparable to traditional therapy. What other problems will AI solve at scale?

This week in Binary Circuit, we study how the frontier AI companies—Claude, OpenAI, and Gemini—are sprinting to outdo one another with potent fresh features. Additionally, according to a ground-breaking Harvard study, AI is breaking down silos and improving performance as it is becoming a real collaborator. We also examine the fast development of humanoid robots and the reasons they might soon change employment and industry as we know it.

Let’s dive in.

The three leading AI platforms—Claude, OpenAI, and Gemini—are engaged in a feature war, each striving to add impressive capabilities.

Recently, Claude has integrated web searches; OpenAI has developed advanced text-to-speech and speech-to-text models; Gemini has launched its Canvas tool.

Anthropic's AI chatbot, Claude, now features real-time web search abilities. This functionality lets Claude access and incorporate current online information with source references.

OpenAI has introduced advanced audio models. Two new speech-to-text models exceed the performance of the previous “Whisper” model, and another novel text-to-speech model offers the capability to respond to stylistic instructions.

These models seek to advance the development of intelligent and adaptable voice assistants. An interactive prototype, OpenAI.fm, enables developers to choose voices and personalize speech features.

The real game-changer? Costs. OpenAI's text-to-speech service, at just $0.015 per minute, is about 85% less than market leader ElevenLabs.

Gemini has launched "Canvas" for group projects. 'Canvas,' an interactive workspace, facilitates real-time collaboration on writing and coding tasks. With Gemini's assistance, users can create initial drafts and refine them by adjusting tone, structure, and content as needed. Canvas provides live previews for coding projects, allowing developers to instantly see the results of their changes.

Your best teammate might not be human. Why business leaders should pay attention.

We’ve all heard about “work husbands” and “work wives.” But what about your cybernetic teammate?

That’s the term coined in a groundbreaking new Harvard study conducted with 776 professionals at Procter & Gamble. The study reveals that AI is not just a tool but a teammate that enhances performance, fosters collaboration, and boosts mood.

Consider this: a machine that aids humans in collaborating more effectively with one another.

Ethan Mollick stated that treating AI as a tool misses the point; it’s evolving into a cybernetic teammate.

Let’s break it down.

AI Can Match—and Sometimes Outperform—Teams

Researchers tested four setups: solo individuals, human-only teams, individuals using AI, and teams using AI.

Here’s what they found:

• Individuals using AI performed as well as human teams—a 0.37 standard deviation boost in performance.

• Teams with AI did even better, outperforming human-only teams and solo AI users.

• AI users completed their work 16.4% faster while writing more and producing higher-quality work.

• Teams using AI were 9.2 percentage points more likely to be in the top 10% of best-performing solutions—3x better than solo humans.

So yes, your AI assistant might be the most productive “employee” on your team.

Breaking Silos and Blending Expertise

Before AI, R&D proposed solutions while sales pitched marketing ideas.

With AI, those lines disappeared. People generated hybrid ideas, balancing technical and commercial aspects regardless of background. AI helped users overcome knowledge gaps and communicate effectively.

AI and the Human Side of Work

Surprisingly, AI users reported more energy, excitement, and enthusiasm while experiencing less stress. The emotional boost from AI matched or exceeded the benefits of team collaboration.

However, despite better performance, AI users felt less confident in their work, a phenomenon termed' AI imposter syndrome”—a discrepancy between performance and self-perception.

What This Means for Organizations

Here’s the takeaway for leaders:

• Don’t just add AI to existing processes—rethink the team itself.

• Use AI to help people contribute outside their silos, whether they’re in marketing, R&D, finance, or ops.

• Recognize that AI-augmented individuals can sometimes do the work of entire teams.

• And maybe, most importantly, train your people to work with AI. Study participants lacked AI expertise, widening the gap between AI haves and have-nots.

This could be the start. As GenAI models improve and employees become comfortable, we’ll likely see greater gains in performance and creativity collaboration.

The future of work isn’t just human vs. machine. It’s likely human + machine.

Share

Humanoid robots could soon be pervasive. What is your strategy to reinvent or build a new business?

Humanoid robots used to seem like science fiction. Now, they’re quietly emerging as one of the fastest-moving—and most transformational—technologies of the decade.

Since the start of 2025, the pace of innovation has gone vertical. Let’s break it down:

A Leap in Capabilities

Boston Dynamics’ Atlas can now pull off cartwheels and breakdancing thanks to reinforcement learning. That’s not just cool—it’s coordination, agility, and control that bring robots closer to handling complex, real-world tasks.

Meanwhile, 1X Technologies is pursuing a more practical approach. Their humanoid, Neo Gamma, is slated for in-home trials in hundreds to thousands of households by the end of 2025. Imagine testing a general-purpose robot in your living room by the year's end.

Tesla aims for 10,000–12,000 Optimus robots this year, scaling to 50,000 by 2026. This isn't a prototype. Tesla is producing 10K+ Optimus robots, aiming for 50K+ in 2026.

Smarter, Cheaper, Faster

NVIDIA launched Isaac GR00T, an open AI foundation model specifically for robots. It’s designed to help humanoids learn faster, generalize better, and move from demo videos to factory floors and city streets.

At the same time, the cost of building high-spec humanoids has dropped by ~40%—from ~$250K to ~$150K in just a year, thanks to broader supply chains and cheaper components.

This drastically shortens the timeline for real-world deployment:

• Factory use cases? A year sooner.

• In-home consumer robots? 2–4 years ahead of schedule.

Market Opportunity is Immense

Goldman Sachs just raised its 2035 market forecast for humanoid robots from $6B to $38B.

Why? Four reasons:

1. A 4x increase in projected shipments (to 1.4 million units)

2. AI acceleration (multimodal and end-to-end learning)

3. Demand from hazardous and aging workforce roles

And that’s just hardware. The labor value of robots could generate $24 trillion in economic output, according to Ark Invest.

From Factory to Front Door

Dr. Peter Diamandis predicts that humanoid robots will begin appearing in homes by 2026—washing dishes, folding laundry, and vacuuming.

A Post-Labor Society?

Diamandis does the math: A $30K robot, leased like a car, could cost $10/day—or about $0.40/hour. That’s a game-changer.

How do you not hire a robot at that rate?

This is especially true when businesses are facing increasing labor shortages, elder care crises, aging workforces, and increasing minimum wages.

We may soon face a profound question: “What kind of world do we want to build in a post-labor society?”

If we do this right, robots don’t just replace labor—they free us from it.

As Sadguru puts it, “Technology is the means by which humanity takes a vacation from basic survival.”

Chart of the week:

The advent and proliferation of AI are driving the rise of revenue per employee. AI is helping companies scale like never before. For instance, Midjourney earns $4.1M per employee, Anysphere generates $3.3M, and OpenAI averages $1.5M with 2,500 workers. The chart below from Digital Native shows the top companies by revenue/employee.

Sound bites you should know:

• Transparent solar has just achieved a record efficiency of 12.3%—without obstructing your view. Building windows could become the new power generators.

• China’s Ant Group reached Nvidia-level AI performance with Alibaba and Huawei chips at a 20% lower cost. China’s quest for self-sufficiency continues to accelerate innovation.

• Aardvark, an AI-driven weather model, delivers accurate forecasts in minutes on a desktop, outperforming traditional systems with only 10% of the data. This is a promising development for better access in developing countries.

Leave a comment

Thank you, for reading!

Cheers,

Team Binary Circuit.

This week in Binary Circuit, we are discussing the key takeaways from NVIDIA’s yearly Global Technology Conference, which showcased new tools that will accelerate the shift toward accelerated computing and AI development and give it a physical shape.

Let’s dive in.

NVIDIA’s GTC provides a glimpse of an AI-driven future. What are the new hardware and software updates, and why do they matter?

NVIDIA's annual GPU Technology Conference (GTC) is a hub for advancements in accelerated computing and AI. This year’s GTC, deemed the "Super Bowl of AI," showcased CEO Jensen Huang's breakthroughs that are set to transform the industry. Described as an "inflection point" in computing, it highlighted advancements in AI performance, open-source tools, hardware breakthroughs, and strategic collaborations reshaping life and business. (watch the keynote here)

This blog post explores key takeaways and their implications for future AI development.

Huang noted that AI's capabilities have reached a turning point. Beyond detecting images and writing text, AI is evolving into an "agentic AI" that can reason and make decisions. Consequently, processing power demand has surged. The company anticipates data center revenue to hit $1 trillion by 2028 to meet GPU demand from major cloud providers.

New Hardware Takes Center Stage. Better GPUs with greater memory are arriving soon, enabling larger AI models and more complex workloads.

The current Blackwell AI platform is in full production, and Blackwell Ultra will arrive in late 2025. The NVIDIA Blackwell Ultra platform connects 36 Grace CPUs and 72 Ultra GPUs. It features 5th-generation Tensor Cores and supports new low-precision types like FP4, enabling over 1,000 TOPS of AI inference per chip.

Huang announced Nvidia's next-gen "Rubin" chips. Named after the astronomer, they are expected to ship next year. The Rubin platform will feature a new CPU and improved GPUs from Nvidia. This new platform has been nearly entirely redesigned for speed and efficiency. These chips are anticipated to enhance performance per dollar, reducing AI processing costs.

Nvidia will release "Rubin Ultra" super chips in 2027. Huang promised better performance in the 2027 Rubin Ultra. According to Nvidia, the forthcoming Rubin Ultra system NVL576 could be 14 times more powerful than top Blackwell systems. This entails adding GPU chips to a server rack and increasing power and cooling.

NVIDIA aims to democratize AI with DGX Spark and DGX Station, termed "personal AI supercomputers." The compact DGX Spark, featuring the GB10 Grace Blackwell Superchip, offers up to 1 PFLOP for model fine-tuning and inference. The Grace-Blackwell architecture enhances the DGX Station, providing data center-level AI capabilities to individual researchers and developers.

NVIDIA announced the RTX Pro Blackwell GPU series for creators and professional workstations. These cards handle demanding AI-driven workflows, 3D design, simulation, and advanced graphics tasks.

Nvidia's Spectrum-X Photonics networking switches enable massive data transfers in AI factories, showing co-packaged optics for the scalability of millions of GPUs. This technology combines electronic circuits with optical pathways, establishing extensive AI infrastructure across multiple locations while reducing energy consumption and costs.

Spectrum-X delivers improvements such as 3.5 times greater power efficiency, 63 times better signal integrity, 10 times enhanced network resiliency, and 1.3 times quicker deployment. These advancements are essential for tackling scalability challenges in building AI infrastructure and ensuring smooth communication and data flow among processors necessary for training complex AI models.

Memory storage needs complete reinvention for AI. Huang pointed out that memory storage is transitioning to semantic-based retrieval systems that can comprehend and deliver knowledge rather than merely raw data. Semantic-based retrieval will facilitate more efficient and intelligent data access for AI workloads, where grasping the context and significance of data is essential. NVIDIA is collaborating with the entire storage industry to provide GPU-accelerated storage stacks to address future AI data storage requirements.

Software and platforms shape the ecosystem of applied intelligence.

NVIDIA paired its hardware advances with a suite of new and improved software tools and platforms designed to accelerate AI development and implementation. Managing AI models and data requires more data centers (or "factories"). Huang saw a future where every company had two factories: one for regular goods and one for AI. AI factories are advanced cloud data centers that teach and offer AI services.

One of the most significant announcements was NVIDIA Dynamo, a new open-source inference tool designed to be the "operating system" for AI factories. Dynamo distributes huge AI reasoning models across hundreds of GPUs, using split serving to boost throughput and reduce latency. NVIDIA claims that Dynamo is the "most efficient solution for scaling test-time compute" and that it can dramatically increase inference throughput while reducing response times and costs.

Another significant breakthrough was the open-source release of cuOpt, NVIDIA's GPU-accelerated decision optimization engine. CuOpt solves tough routing and resource allocation problems in real-time by dynamically examining billions of variables to offer optimal solutions. With major operations research players currently using or considering open-source cuOpt, NVIDIA intends to build an ecosystem around real-time optimization.

Jensen Huang introduced the new open-source reasoning AI models known as the NVIDIA Llama Nemotron family. These models, built on Meta's Llama architecture and post-trained by NVIDIA, excel at tough, multi-step reasoning, code production, and decision-making. NVIDIA's open reasoning model outperforms others, with up to 20% higher accuracy and 5x faster inference.

NVIDIA introduced agentic AI technologies in its AI Enterprise suite to aid intelligent agents. NVIDIA Agent IQ is an open-source toolkit for transparent reasoning and behavior optimization, and the NVIDIA AI-Q Blueprint standard connects knowledge bases to AI agents. New NIM microservices include model serving, continuing education, and real-time adaptation.

NVIDIA launched DGX Cloud Benchmarking, a suite of tools for optimizing AI workloads in the cloud. NVIDIA notes that on popular models like DeepSeek-R1, Llama 3.1 405B, and Llama 3.3 70B on the DGX B200 platform with TensorRT in FP4 precision, they achieve over 3x more inference throughput than the DGX H200 platform.

NVIDIA also showcased the NVIDIA Omniverse Blueprint for AI factories, which includes templates and simulation tools for designing and optimizing huge AI data centers in a virtual environment prior to actual construction.

Additionally, NVIDIA Cosmos provides world-foundation models for robotics training in sophisticated simulation environments.

Robotics is the next big AI frontier.

Huang stated, "The age of generalist robotics is here,” predicting that "physical AI" robotics will be the next trillion-dollar industry. Robots are promising in manufacturing, healthcare, warehouses, and more. Nvidia wants brains and tools for robots.

Isaac GR00T N1, the first open, adjustable foundation AI model for general-purpose humanoid robots, was a category standout. Developers can use this massive AI model (think "brain") to create intelligent robots that can perform numerous functions.

Isaac GR00T N1 will train Nvidia's upgraded Cosmos simulation models to produce humanoid robots, as reported by the Associated Press. Experts believe that this open platform may allow large tech companies, academic researchers, and smaller labs to train robots in new skills using simulated data.

Watch the video here.

Nvidia introduced Cosmos environment models alongside its robot "brain." Cosmos creates complex virtual worlds using Omniverse 3D technology for customizable photorealism. Why does this matter? AI models can learn physics, navigation, and problem-solving by simulating countless environments, like how robots learn by experience.

Huang showcased that Nvidia can generate unlimited, controlled training data with Omniverse and Cosmos. This allows robots to practice in a virtual sandbox more quickly than relying on human-collected data, marking a milestone in synthetic data production for AI training.

Nvidia launched Project Newton, an open-source robotics physics engine. Newton, created with Google DeepMind and Disney Research, will precisely simulate gravity, friction, and object interactions for AI training. This makes virtual-world-trained robots behave authentically in real life. "Blue," a boxy robot, concluded the robotics section with live performances, beeping and booping at Huang. This impressive exhibit showcased the advancements in practical robots.

Other important announcements included:

• Self-driving car partnership: Nvidia partnered with GM to develop autonomous driving and smart factory AI technologies.

• Safety technology for autonomous vehicles: Nvidia unveiled Halos, a full-stack safety technology for autonomous vehicles, covering sensor data processing, driving policy AI, and redundant fail-safes.

• Quantum computing initiatives: Nvidia organized its first “Quantum Day” at GTC 2025, unveiled a hybrid quantum-classical computing platform, and opened a quantum research facility.

What does it mean for the future of AI? Huang's claim that AI compute demand has surged 100x in a year suggests larger, more powerful AI systems. Nvidia’s GTC 2025 reveals a future where AI advances rapidly, driven by powerful GPUs and software.

Smarter AI assistants will take charge of complex workflows, enabling massive gains across economic sectors. Breakthroughs in robotics, cloud computing, and quantum computing aim to democratize AI, making it accessible to businesses and individuals while addressing challenges like energy efficiency and AI ethics.

• Expect smarter AI agents. With higher inference efficiency, conversational AI, real-time translation, and recommendation engines will run faster and cheaper. This means more responsive AI-powered services for customers, such as voice assistants that feel more human and understand context better or streaming services that swiftly customize content. AI will become more common in everyday products and services, even in companies without large AI infrastructures.

• AI workloads could expand 10x or 100x in a few years. Progress is impressive. By the late 2020s, autonomous machines, real-time simulations, and advanced AI models might be common. AI could instantly interpret long discussions or simulate complex scenarios.

• Competitiveness will rise with aggressive progress. Nvidia leads, while AMD and other AI processor giants (Google's TPUs, Amazon's Trainium) are investing to remain competitive. AI hardware competition fosters faster technological advancements, benefiting all. However, if a few corporations dominate the market, it could result in power consolidation. As competition grows, consumers can anticipate stronger, more affordable AI technology.

Energy and infrastructure difficulties arise with so much processing capacity. Nvidia predicts future systems will need more power and cooling. Data centers may need renewable energy and improved cooling technology to develop AI. We must consider sustainability as AI grows.

Chart of the week:

The length of tasks AI agents can complete autonomously has been doubling every 7 months for the past 6 years. A new metric, “task-completion time horizon,” developed by METR, an evaluator of frontier AI models, measures how long it takes AI models to complete tasks at a 50% success rate.

The study predicts that in five years, AI agents will perform many tasks that currently take humans days or weeks. The findings emphasize measuring AI performance by task length to better forecast and understand real-world impact.

Welcome to this week's Binary Circuit. This week, we'll review some incredible technological breakthroughs that are shaping the future. The pace of progress is amazing, with breakthrough interfaces bridging the gap between brain and machine, powerful AI agents automating complex jobs, and innovative energy solutions transforming waste into useful resources. We also discuss the ongoing significance of a distinctively human trait—intuition—in navigating this quickly changing technological and business terrain.

Let’s dive in.

China’s Manus AI agent provides a glimpse into an autonomous future

General-purpose AI agent Manus AI can independently schedule and complete a variety of tasks, including travel booking and stock analysis and execution. Using a multi-agent architecture that functions like an executive supervising specialized sub-agents to break down difficult tasks into reasonable steps, it runs as an AI agent system capable of functioning autonomously and asynchronously.

Manus lays software around current strong models like “Claude Sonnet 3.5” and “Qwen,” not depending on a proprietary model. It can interact successfully with the digital environment by combining several technologies, including web browsers, shell executors, and file reading capabilities.

Early tests and user assessments point to Manus’s remarkable performance—even surpassing OpenAI’s Deep Research on some jobs. From study summarizing and comparative product analysis to creating content and retro-style games, its possible applications are many.

Manus is a powerful sign of where artificial intelligence is headed: a time when more autonomous AI assistants will abound. Industry experts have paid great attention to it and have complimented its capacity to do jobs more efficiently than comparable products like OpenAI's Operator or Anthropic's Computer Use.

The advent of Manus and other AI agents has various significant ramifications, including the automation of employment, concentration on product and engineering, competitiveness and innovation, safety, liability and regulation, impact on the job market, and worldwide competition. Manus AI is still in beta, but its remarkable powers act as a sobering reminder that the age of autonomous AI agents is fast here.

Two breakthrough developments mark a milestone in brain-computer interface research

Rapidly evolving brain-computer interfaces (BCIs) hold great potential to transform human interaction with technology and recover lost capabilities. Recent developments in both non-invasive brain-to-text decoding and two-way adaptive neural interfaces illustrate the fascinating advancement being made in Meta AI as well as in Tianjin University and Tsinghua University in China.

Designed to decode sentences from brain activity recorded by electroencephalography (EEG) and magnetoencephalography (MEG), Meta’s Brain2Qwerty, a deep learning architecture, closes the performance gap between non-invasive and invasive BCI approaches.

MEG outperformed EEG with a striking 32% average character-error-rate (CER) compared to EEG's 67%. Top MEG participants achieved an amazing 19% CER, with the model able to decode phrases beyond the training set.

It demonstrates that non-invasive BCIs can approach the accuracy of invasive implants and enable safer communication solutions for non-communicative patients.

China's researchers have revealed the first two-way adaptive brain-computer interface (BCI) ever published, using a memristor-based neuromorphic decoder to achieve a 100-fold boost in efficiency and four degrees of freedom in movement control. The system’s dual-loop feedback method consists of a user feedback loop enabling users to gradually refine their thought patterns over time, hence boosting control and flexibility, and a machine learning loop continuously updating the brain wave decoder. This mutual adaptation between human cognition and machine intelligence is a basic first step toward brain-computer co-evolution.

This BCI's improved control features highlight its accuracy by enabling the simple operation of virtual drones with precisely tuned motions. It also promises flawless integration with external devices driven by artificial intelligence, hence improving user experience.

The adaptive character of the system makes the lower cognitive load possible, thereby making the system perhaps accessible to people with different cognitive capacities. These developments open the path for complex uses, including enhanced prostheses, human-machine symbiosis, and medical rehabilitation.

These discoveries complement other global BCI advances including Neuralink's implanted BCIs for thought-controlled device control and Precision Neuroscience's minimally invasive BCIs. China's regulatory developments seek to hasten the growth and support high-quality BCI industry innovations.

Share Binary Circuit

Microsoft's Magma enables intelligent AI agents to operate in digital and physical environments

Microsoft Research introduced Magma, a foundation model for multimodal AI agents that perceives the world and performs goal-driven tasks across digital and physical contexts. Magma addresses the limitations of existing AI models in adapting to various environments.

Magma seeks to be a single foundation model able to adapt to new activities in both digital and physical environments, therefore allowing AI-powered assistants and robots to see their surroundings and propose suitable actions.

Magma uses Set-of-Mark (SoM) to designate actionable visual items in images (like clickable buttons) and Trace-of-Mark (ToM) to monitor object movements in movies. These methods teach Magma spatial intelligence from enormous training data and help create meaningful actions from visual and verbal inputs. Tokenized texts and shared-vision encoded images/videos are input into a large language model (LLM) to generate verbal, spatial, and action outputs.

Magma can improve AI assistants for step-by-step navigation and UI automation. A deeper understanding of real-world objects can lead to more adaptable robots. From healthcare to industrial automation, Magma’s multimodal intelligence has the potential to redefine collaboration between humans and AI interaction.

Magma is a major step towards a future when AI really understands and interacts with the complexity of our environment.

Batteries from nuclear waste could solve many challenges in digitally connecting remote assets

Ohio State University has created a revolutionary nuclear battery that can generate electricity using radioactive atomic waste products. This creative technique can power small appliances while solving the major problems with nuclear waste management.

Like rooftop solar panels, the battery runs on solar cells (photovoltaic cells), which translate light into electricity, and scintillator crystals, which are uniquely able to emit light when subjected to radiation.

The study team tested its prototype, which is roughly the size of a four-cubic-centimeter cube, using two typical radioactive isotopes present in spent nuclear waste: cesium-137 and cobalt-60. The findings were hopeful: The battery developed 288 nanowatts of electricity when run on cesium-137.

Especially in environments like nuclear waste storage facilities, deep-sea research, and even space missions where high levels of radiation are already present, this technology offers a potential way to transform a problem into a useful resource. Importantly, the battery is safe to handle even though it utilizes surrounding gamma radiation, as it does not contain any radioactive materials.

Before the technique can be generally adopted, though, various difficulties still exist, including growing power output, minimization, durability, safety, and regulatory approval. Notwithstanding these obstacles, the researchers have great hope for the future. They think the nuclear battery idea will establish a significant place in the sensing sector as well as energy generation.

The development of a nuclear waste-powered battery is an important step toward sustainable nuclear energy. This innovation offers long-lasting, maintenance-free power, showing that one person's waste can be another's resource.

Intuition, often regarded as a gut feeling, provides leaders with a crucial advantage in a rapidly changing world

Intuition is a form of subconscious processing grounded in our past experiences and insights, not a magical phenomenon. Science indicates that the brain unconsciously processes knowledge far faster than our conscious thinking, resulting in more informed decisions.

Studies have indicated that intuition is especially useful when handling complicated data or making snap judgments. People who trusted their intuition experienced more life happiness, inventiveness, and self-awareness; they also solved challenging tasks faster and more precisely. Moreover, those performing well on intuition tests have been demonstrated to make better investing choices.

Historical figures like Phil Knight, the creator of Nike, Jensen Huang of NVIDIA, Steve Jobs, Oprah Winfrey, and Albert Einstein have all acknowledged their intuition for watershed moments.

System 1 Thinking, which is fast, instinctive, and emotional—the domain of gut feelings and snap judgments—has become increasingly important in today's fast-changing AI world. Leaders negotiating AI-driven sectors must foresee changes before they become clear and react quickly, depending on their natural awareness of the terrain.

Conscious activities such as cultivating presence, listening to your body, practicing self-reflection, trusting the universe, and believing in your first impressions will help you develop your intuitive edge.

When others hesitate, visionary leaders often take big risks, knowing that waiting for perfect conditions or complete data could mean losing important chances. Leaders who learn to maximize and trust their natural edge will be able to negotiate uncertainty, inspire creativity, and finally propel transforming success.

Chart of the week:

How soon will AI augmentation give way to complete automation?

AI is currently used to support professionals rather than replace them.

The Anthropic Economic Index shows that 57% of AI usage is for augmentation, where AI assists humans with tasks. Complete automation, where AI handles tasks with minimal human input, is at 43%.

AI agents’ quick development for niche applications at a lower cost could soon swap the equation.

Sound bites you should know:

Google's newly launched open-source AI, Gemma 3, is designed for efficiency, boasting a 128K context window and the ability to run on a single GPU. It can even run locally on a smartphone.

Robot-assisted surgeries are on the rise! 22% of surgeries now use robots, offering less bleeding, pain, and recovery time. Advances in surgery are coming.

OpenAI's new platform enables companies to boost productivity by creating custom AI bots for different business uses. AI agents at a factory scale?

Google's Gemini Robotics uses AI to let robots comprehend, interact with, and manage things like humans. It controls industrial arms and humanoid assistants using vision, language, and action, a crucial step toward general-purpose robots. AI is becoming physical.

Microplastic pollution is damaging plants’ ability to photosynthesize, leading to crop loss and potential harm to the planet. What's the true cost of our plastic addiction?