100 Leaders Shaping the Future of Artificial Intelligence

2. Jensen Huang ' to the clipboard">

• Founder & CEO, Nvidia

Jensen Huang has leveraged Nvidia’s A.I. infrastructure dominance to reshape global A.I. development and policy. As co-founder and CEO since 1993, Huang has transformed Nvidia from a company making processors for video games into the backbone of the A.I. revolution. Nvidia’s GPUs, first introduced in 1999, now power the majority of modern A.I. systems—and, this summer, helped Nvidia become the world’s first public company to reach $4 trillion in market value.

In recent months, the executive has also overseen various product releases at Nvidia geared towards physical A.I. systems like robotics and self-driving cars as the chipmaker eyes a future beyond GPUs. “We stopped thinking of ourselves as a chip company long ago,” said Huang in June.

Huang’s influence extends into geopolitical A.I. policy. In July, he persuaded President Donald Trump to reverse bans on A.I. chip sales to China and agreed to give the U.S. government 15 percent of Nvidia’s revenue from such sales in return, demonstrating his ability to shape international A.I. trade relationships amid ongoing security concerns over semiconductor exports to China. His appointment in April to the U.S. Department of Homeland Security’s AI Safety and Security board positions him at the center of national A.I. security policy development.

Beyond policy influence, Huang drives A.I. advancement through strategic investments in research infrastructure. He contributed $50 million to Oregon State University for a $200 million supercomputing institute and $30 million to Stanford University to establish the Jen-Hsun Huang School of Engineering Center. His recognitions include the VinFuture Grand Prize and the 2025 Edison Achievement Award. This combination of market leadership, policy influence and strategic research investment establishes Huang as a key architect of global A.I. infrastructure and governance frameworks.

5. Elon Musk

• Founder, xAI

Best known for leading Tesla, SpaceX and X (formerly Twitter), Elon Musk added another disruptive venture to his empire in July 2023 with the launch of xAI, the generative A.I. company behind the Grok chatbot. In less than two years, xAI has vaulted into the front ranks of the industry, powered by Musk’s characteristic mix of audacious vision, aggressive financing and headline-grabbing bets on infrastructure.

Last year, xAI unveiled Colossus, a Memphis-based supercomputing facility billed as the world’s largest. Musk has framed Colossus as essential to accelerating A.I. toward human-level reasoning, though the project has drawn criticism for its massive environmental footprint and water demands. Musk has since floated plans for a wastewater recycling facility to ease community pushback.

xAI’s rapid expansion has included a $33 billion acquisition of X and the purchase of Hotshot, an A.I. video-generation platform. To fuel its growth, the company has raised billions—$6 billion in equity funding led by Andreessen Horowitz and BlackRock, plus a $10 billion debt and equity deal in mid-2025. In July, xAI secured a $200 million U.S. Department of Defense contract and rolled out Grok 4, a model that scored at state-of-the-art levels on ARC-AGI-2 benchmarks. Musk has touted Grok 4 as a step toward A.I. capable of discovering new technologies and even “new physics.” Yet, its permissive “free speech” standards have generated backlash, with critics citing antisemitic, sexually explicit and violent outputs.

Musk has infused A.I. directly into his other companies. Tesla relies on advanced neural networks and self-supervised learning systems to power its Autopilot and Full Self-Driving software. Meanwhile, X has integrated Grok across its social media platform, offering subscribers conversational A.I. features and experimenting with automated content moderation, recommendation systems and A.I.-generated media—all part of Musk’s broader push to turn the platform into an “everything app.”

Musk has also set his sights on talent wars. Earlier this month, he announced plans for a Seattle engineering hub, widely seen as an attempt to siphon talent from nearby rivals such as Microsoft, even as Musk sues the company over alleged anti-competitive ties with OpenAI.

7. Alexandr Wang

• Chief A.I. Officer, Meta

At just 28, Alexandr Wang has become one of the most consequential figures shaping the trajectory of A.I. This summer, the founder and former CEO of Scale AI joined Meta to lead its newly formed Meta Superintelligence Labs (MSL) after Meta acquired a 49 percent stake in Scale for $14.3 billion. It marks one of the largest corporate realignments in the A.I. industry to date.

MSL will serve as an umbrella organization overseeing four separate A.I. units at Meta. In addition to A.I. products, infrastructure and long-term research, the division includes TBD Labs, a group stacked with new hires from rivals like OpenAI, Google and Anthropic that is focused on achieving superintelligence, a form of A.I. with capabilities exceeding that of humans. Wang’s meteoric rise has been defined by his commitment to scale. Since launching Scale in 2016, he has built the company into a backbone of enterprise and government A.I. deployment. Scale’s core business is preparing and organizing the massive datasets that train A.I. systems—turning raw information like images, video, text and sensor data into structured inputs machines can learn from. Over time, the company has expanded into building infrastructure that helps organizations test, fine-tune and securely deploy A.I. models. 

Wang has also become a sought-after presence on global stages from Davos to the World Economic Forum, where his views on the responsible deployment of A.I. resonate with policymakers, corporate leaders and researchers alike.

Now, with his new role at Meta, Wang has evolved from entrepreneur to strategist at the nexus of corporate power and frontier research. His influence extends beyond technical infrastructure into shaping how advanced A.I. will be built, governed and integrated into society. As the race toward artificial general intelligence accelerates, Wang’s trajectory suggests he will remain pivotal in guiding not only how A.I. develops, but how the world learns to live with it.

9. Satya Nadella

• CEO & Chairman, Microsoft

Satya Nadella helped Microsoft enter the A.I. race early, leading the tech giant’s initial investment into OpenAI, then a little-known research lab, in 2019. That bet proved prescient. The partnership has since grown into a $13 billion alliance, giving Microsoft exclusive access to OpenAI’s most advanced models while positioning Azure as the default infrastructure for their deployment. Nadella has expanded this strategy by backing other A.I. developers, including France’s Mistral AI, to diversify Microsoft’s capabilities further.

Beyond high-profile partnerships, Nadella has aggressively pushed to build Microsoft’s in-house A.I. systems. In March 2024, Nadella boldly hired Mustafa Suleyman, co-founder of DeepMind and Inflection AI, to lead Microsoft’s consumer A.I. business. The appointment underscored Nadella’s vision of putting A.I. at the center of Microsoft’s consumer-facing products, from Copilot in Office and Windows to Bing search. Bringing in one of the field’s most respected entrepreneurs also signaled Nadella’s ambition to stay ahead in the global A.I. race.

In August, Microsoft unveiled its first in-house models: MAI-Voice-1, a speech system, and MAI-1-preview, designed for everyday queries. Earlier this month, Microsoft told employees it would significantly expand investments in physical A.I. infrastructure to make its models more competitive. On September 16, it announced a more than $30 billion commitment over four years to grow U.K. operations and A.I. infrastructure.

Internally, Nadella has acknowledged A.I.’s disruptive potential, preparing teams for structural shifts as automation changes how work gets done. Under his leadership, Microsoft has gone from a software darling to A.I. behemoth, where every product—from Office to Azure—is being reimagined with machine learning at the core.

11. Sundar Pichai

• CEO, Google & Alphabet Inc.

Under Sundar Pichai’s leadership since 2015, Google has established itself as a global A.I. powerhouse. Its DeepMind unit produced the groundbreaking AlphaGo, its researchers pioneered transformer-based models through the pivotal Attention Is All You Need paper, and three of last year’s Nobel Prize winners either worked at Google or trained within its ranks. Today, the company’s flagship A.I. assistant Gemini counts more than 450 monthly users, while its A.I.-generated summaries reach over 2 billion customers.

Much of that growth stems from the global rollout of AI Mode, an agentic search feature now serving more than 100 million monthly active users in the U.S. and India. Powered by Gemini 2.5, the tool enables conversational, multi-step queries that bring A.I. reasoning directly into Search.

Some of the company’s more recent A.I. updates include adding agentic features to A.I. mode, expanding accessibility for Gemini Deep Think—a reasoning model that won this year’s gold medal at the International Mathematical Olympiad (IMO)— unveiling new image editing models, and releasing the world model Genie 3. Like its Silicon Valley rivals, Google is also making progress on developing A.I.-powered smart glasses through partnerships with Warby Parker, Gentle Monster and Samsung.

To sustain these ambitions, Google plans to spend $85 billion this year, funding data centers, servers, networking and talent. This summer, Pichai brought on Varun Mohan and Douglas Chen, the CEO and co-founder of A.I. software firm Windsurf, to strengthen Google’s agentic coding tools. With Pichai’s conviction that A.I. will reshape the future of computing, the company shows no signs of slowing its pursuit of the next breakthroughs.

13. Mira Murati

• Founder, Thinking Machines Lab

Mira Murati, who previously served as CTO of OpenAI, had a brief stint as the ChatGPT-maker’s CEO and held a senior engineering role at Tesla, co-founded Thinking Machines Lab in late 2024 to address key gaps in understanding foundation models and other machine learning systems. Within five months of launching, Murati secured $2 billion in seed funding from investors like Andreessen Horowitz, Nvidia and Jane Street, bringing the company’s valuation to $12 billion.

Last week, Thinking Machines Lab published the research behind one of its first projects, investigating why so many A.I. models produce inconsistent or seemingly random results. The post appeared on the company’s new blog, Connectionism, and suggested solutions to boost LLM determinism. Murati said in July that the startup plans to unveil its first product later this year, hinting that it will include an open-source component and support researchers and startups as they develop custom models.

With a team drawn from OpenAI, Character.ai, Mistral and other top A.I. companies, Thinking Machines Lab has become highly sought-after talent—Meta reportedly tried to poach a dozen employees with offers totaling more than $1 billion, but none accepted. Murati emphasizes the collective nature of scientific progress and, rather than focusing on fully autonomous A.I. systems, aims “to build multimodal systems that work with people collaboratively” and distribute her startup’s research across the broader A.I. community.

15. Reid Hoffman

• Co-Founder & Board Chair, Manas AI

Reid Hoffman may be best known as the co-founder of LinkedIn, but in recent years, he has also become one of A.I.’s most prominent backers, investing in startups spanning communication, healthcare and beyond. In January, he doubled down on his optimism with the book Superagency, co-authored with tech writer Greg Beato, arguing that A.I. will enhance rather than diminish human agency.

What strikes Hoffman most is how quickly A.I. has entered the mainstream, adopted even by people outside the tech world for everything from managing relationships to boosting workplace productivity. “I don’t recall a similar level of wide-ranging mainstream adoption just a couple of years into the Web 1.0 era,” he tells Observer. “As much as there are obvious concerns and even resistance to these new technologies, the public’s overall embrace has been eye-opening.”

In 2022, Hoffman co-founded Inflection AI, which was effectively absorbed by Microsoft last year. More recently, his focus has shifted toward healthcare. In June, he led a $12 million funding round for Sanmai Technologies, a startup pairing A.I. with ultrasound. And this year, he helped launch Manas AI, an A.I.-powered drug discovery company that raised $24.6 million in seed funding from Hoffman, General Catalyst and Greylock Partners. The venture aims to accelerate and lower the cost of developing cancer and rare disease treatments through generative computational chemistry, biology and proprietary A.I. systems.

Manas AI’s founding team reflects that ambition. Hoffman launched the company with Siddhartha Mukherjee, the Pulitzer Prize–winning cancer physician and author of The Emperor of All Maladies. Jonathan Baell, a renowned medical chemist with more than two decades of experience, serves as chief scientific officer.

Hoffman tells Observer that he’s confident widespread user adoption, not investor hype, is driving A.I.’s impact. But that doesn’t mean the technology’s popularity doesn’t come with drawbacks, as repetitive use of learning-based A.I. systems could widen the gap between leading A.I. models and emerging competitors. Such concentration is “concerning on multiple levels,” Hoffman says, noting that early design choices and development contexts will shape A.I.’s trajectory and whose interests it ultimately serves. “It could lead to a world where we see a decline in the kind of entrepreneurial dynamism that has traditionally been the source of new job creation, competition, innovation and broad-based productivity growth.”   

17. Marc Andreessen

• Founder, Andreesen Horowitz

Marc Andreessen’s reputation as one of Silicon Valley’s most aggressive backers of A.I. ventures has only grown stronger in 2025. In July, his venture capital firm Andreessen Horowitz (a16z) invested $2 billion into Thinking Machines Lab, the new startup founded by former OpenAI CTO Mira Murati. The deal valued the company at $12 billion before it had even launched a product—underscoring Andreessen’s conviction in betting early on high-profile A.I. talent.

A16z is also in early discussions to raise a $20 billion fund dedicated to growth-stage A.I. startups, with commitments to major portfolio companies like Databricks, Elon Musk’s xAI and Mistral. According to Crunchbase, the firm was the most active post-seed investor in A.I. during 2024, completing 42 deals and outpacing nearly every other venture capital competitor.

To support its rapidly expanding portfolio, a16z launched Oxygen last October, an initiative that gives startups access to more than 20,000 Nvidia GPUs. The program not only addresses the acute shortage of computing power but also cements Andreessen’s influence over the infrastructure that fuels cutting-edge A.I. research. His broader vision is equally bold: scaling U.S.-based factories for robotics and A.I., making open-source models the global standard, and preparing for a future in which A.I. replaces nearly every job—except, as he often quips, his own.

Beyond capital deployment, Andreessen has moved directly into shaping A.I. policy. In August, he announced the formation of Leading the Future, a new political action committee into which he and other industry leaders have committed $100 million. The PAC is designed to advance pro-innovation regulatory frameworks in Washington and counterbalance growing calls for restrictive oversight.

19. Joelle Pineau

• Chief A.I. Officer, Cohere

Joelle Pineau’s move from leading Meta’s FAIR research division to shaping Cohere’s A.I. strategy marks a pivotal moment in enterprise A.I. After nearly eight years at Meta, where she championed open-source projects like Llama and PyTorch, Pineau stepped down in April. Four months later, Cohere closed a $500 million round led by Radical Ventures and Inovia Capital, with backing from AMD, Nvidia, PSP Investments and Salesforce Ventures, lifting its valuation to $6.8 billion. The funding coincided with Pineau’s appointment as chief A.I. officer. Pineau disputes the notion that A.I. is an opaque “black box.” While complex, she argues, enterprise models can often be traced more clearly than human reasoning. “It’s not impossible to understand how a prompt leads to an output,” she tells Observer. 

Her focus at Cohere diverges from rivals chasing artificial general intelligence. “The approach Cohere is taking is more focused than players that are chasing AGI or general superintelligence, and that gives us a leg up in the enterprise market,” she explains. By prioritizing privacy, security and compatibility with sensitive enterprise data, Cohere is positioning itself in critical sectors like finance, healthcare, telecom and government.

Pineau also applies computer security principles to A.I. development, contending that “open protocols are in fact more secure, because flaws are discovered much faster and properties are better understood.” This philosophy underpins Cohere’s open science model, even as she stresses that “enterprises can’t afford to have data leak—whether it’s internal proprietary data or sensitive customer data.”

The recent boom in A.I.-assisted software development has particularly impressed her. LLMs, she notes, now accelerate code generation, bug fixes and developer support while enabling “almost anyone, even with very little computer science training, to implement their ideas quickly.” Such advances also open the door to A.I. systems that can self-improve. Her vision emphasizes building systems that are “traceable, controllable and customizable,” with rigorous evaluations that consider both technical performance and broader impacts like bias, transparency and safety.

Pineau joins Cohere at a time of rapid expansion. Its North platform, built for secure A.I. agents, is now complemented by tailored enterprise models and partnerships with Oracle, Dell, LG and RBC. Her arrival signals Cohere’s intent to fuse ethical research with scalable impact in the enterprise A.I. market.

21. Lisa Su

• CEO, AMD

Lisa Su is transforming AMD from a niche chipmaker into a crucial player in the A.I. era, as Big Tech races to secure powerful semiconductors to fuel energy-hungry data centers.  A Taiwanese-born American engineer with a Ph.D. in electrical engineering from MIT, Su became AMD’s CEO in 2014, when the company was struggling to stay afloat. By focusing on high-performance computing and doubling down on semiconductor innovation, she orchestrated one of the most remarkable turnarounds in tech history, positioning AMD as a rival to giants like Nvidia and Intel. 

Under Su’s leadership, AMD has become a critical supplier of chips powering cloud computing, gaming and increasingly A.I. Su has predicted that the global demand for A.I. chips could rapidly increase to more than $500 billion in a few years. At AMD’s Advancing A.I. event in June, Lisa Su introduced the Instinct MI350 series, a family of GPUs that delivers four times the computing power and 35 times better inference performance than previous generations. She also previewed Helios, a rack-scale A.I. system designed for large-scale training and deployment, set to launch in 2026. Looking further ahead, at a recent investor conference, AMD executives said the company’s next-generation chip, MI450, will outperform any rival hardware, including Nvidia’s Rubin Ultra. 

Su’s vision extends beyond hardware. She has championed ROCm, AMD’s open-source software stack, which provides developers with a flexible, efficient platform to build advanced A.I. applications. Su has been widely recognized as one of the most influential executives in technology and is the first woman to receive the IEEE Robert N. Noyce Medal, awarded for outstanding contributions to the microelectronics industry.

23. Stéphane Bancel

• Founding CEO, Moderna

Stéphane Bancel represents one of the most radical artificial intelligence integrations in pharmaceutical development, transforming Moderna from a biotech company into a laboratory for A.I.-driven organizational evolution. Since partnering with OpenAI in 2023, Bancel has deployed over 3,000 custom GPTs across Moderna’s operations, with employees averaging 120 weekly ChatGPT Enterprise conversations. His ambitious mandate: every employee should use ChatGPT at least 20 times daily.

In May, Bancel merged Moderna’s HR and Technology departments under a single leader, reflecting his belief that A.I. fundamentally changes how work is structured and performed. This organizational revolution positions Moderna as a living experiment in A.I.-native corporate structure, with specialized A.I. tools like “Dose ID GPT” automating clinical trial dose optimization and the legal team achieving 100 percent ChatGPT Enterprise adoption. Bancel claims that while launching 15 new mRNA products in five years would traditionally require 100,000 employees, Moderna can achieve this with its current 6,000 staff through A.I. leverage. His bold prediction extends even further: with A.I. assistance, scientists will “understand most diseases” within 3-5 years, representing “a first for humans on the planet.”

However, this vision comes with difficult realities. In July, Bancel announced plans to reduce Moderna’s workforce by 10 percent (approximately 580 employees) by the end of 2025 while targeting $1 billion in cost reductions. Despite A.I. investments, Moderna’s 2025 revenue guidance of $1.5-2.5 billion represents a significant decline from 2024’s $3.0-3.1 billion, highlighting the complex challenge of balancing A.I. transformation with commercial performance. Bancel’s willingness to reduce the traditional workforce while massively scaling A.I. deployment demonstrates the potential and the challenges that define the current A.I. era. His approach provides a blueprint for how artificial intelligence can transform entire industries, not just individual functions.

“We believe very profoundly at Moderna that ChatGPT and what OpenAI is doing is going to change the world,” Bancel has said. “We’re looking at every business process and thinking about how to redesign them with AI.” While other CEOs discuss A.I. adoption, Bancel implements it at unprecedented depth.

25. Doug McMillon

• CEO, Walmart

For over a decade, Walmart CEO Doug McMillon has guided the retail giant through an A.I. transformation that touches nearly every corner of its business. Under his leadership, the company has prepared more than 850 million data points for A.I. training, rolled out its first A.I.-powered merchant assistant, “Wally,” and added senior roles dedicated to accelerating adoption—including an EVP of A.I. Acceleration, Product and Design, and an EVP of A.I. Platforms.

This year, Walmart unveiled a suite of new agents designed for stakeholders across its ecosystem. “Marty” serves sellers, “Sparky” supports customers, and a “Super Agent” provides associates with a single point of entry to all A.I. tools. The company also introduced a “Developer Agent” to streamline software launches. Together, these systems aim to boost efficiency while standardizing how Walmart employees, partners and shoppers interact with the company’s digital infrastructure.

Walmart is also applying A.I. to its vast physical footprint. The retailer can optimize truck load packing using digital twins and “detect, diagnose and remediate issues” before they disrupt operations. The impact has been tangible: in 2024, Walmart reported a 30 percent drop in global emergency alerts and a 19 percent reduction in refrigeration maintenance spending across its U.S. stores thanks to digital twin technology.

McMillon has been vocal about Walmart’s shift from a retail company to a technology leader. “We are really good when it comes to technology these days,” he said earlier this year, underscoring his belief that A.I. is central to Walmart’s long-term competitiveness.

Earlier this month, Walmart announced a partnership with OpenAI to design an A.I. certification program for its workforce, slated to debut in 2026. The initiative signals Walmart’s intent not only to adopt cutting-edge tools, but also to upskill its employees at scale, ensuring its 2.1 million associates worldwide are prepared for the next wave of A.I.-driven retail.

27. Peter Thiel

• Managing Partner, The Founders Fund

Peter Thiel’s empire shows no sign of slowing, largely thanks to his high-stakes bets on A.I. This year, his venture capital firm Founders Fund closed a $4.6 billion growth fund, underscoring renewed confidence in frontier innovation spanning A.I., defense and advanced manufacturing. One of its boldest moves came in a $2.5 billion round in Anduril, which develops autonomous systems for military use. The round pushed the defense-tech company’s valuation to $31 billion.

Thiel’s fingerprints are on some of the most influential players in the field. Founders Fund was an early backer of OpenAI and Scale AI. This year, Founders Fund added another major stake, leading a $400 million round for Cognition AI—the San Francisco startup behind Devin, an A.I. software engineer now in use at Goldman Sachs—doubling its valuation to $10.2 billion.

Thiel co-founded Palantir Technologies, the controversial data analytics company that creates sophisticated software platforms to help government agencies and corporations operationalize vast amounts of disparate data. Palantir provides what former employees describe as “extravagant plumbing with data,” and positions itself as selling not just software but “the idea of a seamless, almost magical solution to complex problems” to governments and Fortune 500 companies. In August, the company reported it had surpassed $1 billion in quarterly revenue for the first time.

Despite his sweeping portfolio, Thiel maintains a contrarian pragmatism about A.I.’s impact. On the Interesting Times podcast, he described the technology as “more than a nothing burger” but “less than the total transformation of our society.” For Thiel, A.I. is not an end in itself but a bulwark against what he calls technological stagnation. Without breakthroughs in areas like space travel or cures for dementia, he warns, innovation risks sputtering out.

29. Jonathan Ross

• Founder & CEO, Groq

Jonathan Ross is positioning Groq as a formidable challenger in the A.I. chip market with its language processing units (LPUs), designed to dramatically boost compute speed while cutting costs. His philosophy is simple: the market needs both speed and accuracy. “While GPUs could train large models, they were expensive and slow to run,” Ross tells Observer. Groq’s LPUs, by contrast, can run trillion-parameter models quickly and cost-effectively. For Ross, the challenge is existential: “You can only run A.I. if you have enough inference compute. It’s like cars; you need oil to drive.” He warns that demand for inference compute is outpacing global manufacturing capacity—a trend that could concentrate A.I.’s benefits among a few.

Over the past six months, Ross has aggressively expanded Groq’s footprint through high-profile partnerships. In February, he secured $1.5 billion from Saudi Arabia to grow its A.I. infrastructure, including Groq’s Dammam data center, a deal expected to generate $500 million in revenue this year. In April, Groq partnered with Meta to give developers the fastest, most cost-effective way to run Llama 4 models. May brought an exclusive partnership with Bell Canada to power Canada’s largest sovereign A.I. infrastructure project, alongside new data centers in Houston and Dallas. In July, Groq opened its first European data center in Helsinki, Finland, in partnership with Equinix, and plans to announce its first Asia-Pacific location later this year. By August, Ross unveiled a collaboration with OpenAI to provide developers instant access to openly licensed models at scale. 

Despite recently cutting 2025 revenue projections from $2 billion to $500 million, Groq closed a $750 million fundraising round this week, at a $6.9 billion valuation. Ross compares the industry to oil exploration: “If you measure by attempts, most A.I. projects fail. If you measure by spend, most yield positive returns. Some A.I. ‘strikes’ have been very profitable—more than the total investment to date.”

For Ross, A.I.’s accelerating pace is undeniable. In one client meeting, he requested a new feature from his engineers. Hours later, A.I. had built and deployed it. “The next phase shift will be when that can happen before the meeting ends,” he says.

31-33. Nick Frosst, Aidan Gomez & Ivan Zhang

• Founders, Cohere

Following a $500 million funding round in August (led by Radical Ventures and Inovia Capital), Nick Frosst, Aidan Gomez and Ivan Zhang’s Cohere reached a $6.8 billion valuation, establishing the Toronto-based startup as one of the world’s most promising A.I. companies. Since December 2024, Cohere has released multiple flagship models demonstrating technical superiority. Rerank 3.5, released in December, delivers 26.4 percent improvement on cross-lingual search and 23.4 percent better performance than Hybrid Search on financial services datasets across 100+ languages. Secure A.I. agents platform Cohere North launched in January, outperforming Microsoft Copilot and Google Vertex AI Agent Builder across multiple benchmarks. Command A, released in March, operates on just 2 GPUs (versus competitors requiring 32) and achieves 1.75x faster token generation than GPT-4o. And with April’s release of Embed 4, Cohere introduced a multimodal embedding model enabling search across complex PDF reports and presentations with text, images, tables, graphs and diagrams across 100+ languages.

An efficiency-first approach reflects the founders’ contrarian philosophy. “The industry got obsessed with throwing more money and chips, leading to better outcomes, but we’ve proven that wrong repeatedly,” Zhang tells Observer. Frosst also advocates for A.I. efficiency over scale, consistently arguing against the “bigger is better” approach. This position proved prescient with recent market shifts toward more efficient models. 

Under their leadership, Cohere more than doubled its annualized revenue from $35 million in March to over $100 million by May this year, thanks mainly to its partnerships with governments, enterprises and specialized sectors.

“We knew enterprise adoption would be slower than consumer, but we’re getting to the point where people realize this isn’t just another productivity tool,” Zhang says. “It’s not experiential anymore; it’s becoming real infrastructure.”

Government partnerships include operational transformation and A.I. safety research with the Canadian AI Safety Institute (CAISI) and the U.K.’s AI Security Institute (AISI), plus Second Front for secure U.S. and allied government A.I. via the 2F Game Warden platform. The startup also recently agreed to examine how its technology can boost public service operations across Canada’s federal government. 

Major enterprise deployments include Oracle powering 200+ A.I. features across NetSuite for tens of thousands of enterprises, SAP for comprehensive global A.I. integration, Dell as the first on-premises North provider, and co-developing North for Banking with Canada’s largest bank, RBC. Regional expansion covers Japan through Fujitsu’s exclusive partnership with co-developed Takane LLM, South Korea via LG CNS, the Middle East through stc Group’s $83B telecom partnership, and Canada with Bell’s sovereign A.I. Fabric solution.

As Zhang explains, Cohere’s enterprise focus stems from simple math. “Enterprises will pay for A.I. that solves their actual business problems. You can’t just take a general consumer model and expect it to work in a regulated environment,” he says. Rather than chasing “flashiest demos,” the team built “infrastructure that works when you need to deploy at scale with real stakes.”

35. Simon Kohl

• Founder & CEO, Latent Labs

From Nobel Prize-winning protein prediction to breakthrough drug design, Simon Kohl has positioned himself at the forefront of A.I.’s transformation of biology. He co-led Google DeepMind’s protein design team and was a senior research scientist on DeepMind’s AlphaFold2, the project that earned Demis Hassabis and John Jumper the 2024 Nobel Prize in Chemistry. “Having co-developed AlphaFold2, I’ve seen firsthand how A.I. can solve incredibly complex problems,” Kohl tells Observer.

Before leaving DeepMind, Kohl built AlphaFold2’s widely used uncertainty prediction system “pLDDT” and set up DeepMind’s wet lab at London’s Francis Crick Institute. When Kohl realized it was possible to “move beyond just predicting biological structures to actually designing them from scratch,” Kohl decided to found Latent Labs. “We were at an inflection point where generative A.I. could make biology programmable.” In 2024, Latent Labs was one of the early-stage startups to receive support from AWS through its Generative A.I. Accelerator. 

In February, Latent Labs raised $50 million in venture capital funding. Angel investors include Google Chief Scientist Jeff Dean, Cohere founder Aidan Gomez and ElevenLabs founder Mati Staniszewski. 

In July, the company launched LatentX, achieving 91 percent to 100 percent hit rates for macrocycles and 10 percent to 64 percent for mini-binders across seven therapeutic targets in wet lab experiments. Unlike traditional methods that predict existing structures, LatentX simultaneously designs the molecular sequence and 3D structure of proteins in real-time, following atomic-level physical rules to create entirely novel molecules. “We’re not just understanding nature anymore, we’re becoming capable of authoring it with precision,” he says. “Scientists achieve in 30 candidates what previously required testing millions, turning months of experiments into seconds of computation.” Traditional drug discovery hit rates are typically below 1 percent, Kohl explains.

Bringing experience from DeepMind, Microsoft, Google, Stability AI, Exscientia, Mammoth Bio, Altos Labs and Zymergen, Kohl’s team is prioritizing oncology, autoimmune diseases and rare genetic disorders—areas where conventional drug discovery faces significant challenges. The company is particularly focused on macrocycles, which combine the precision of biologics with the oral deliverability of small molecules. In direct laboratory comparisons, Latent Labs has outperformed results from major technology companies and leading academic institutions, leveraging their team’s AlphaFold experience combined with enterprise-grade platform engineering.

Rather than developing proprietary medicines, Latent Labs licenses its technology through a web-based platform, making advanced A.I. accessible to academic institutions, biotech startups and pharmaceutical companies. While making the technology broadly accessible, Latent Labs maintains strict biosafety protocols, actively engages with regulators on dual-use concerns, and validates all computational designs in its physical laboratory to ensure real-world safety.

“We envision a future where effective therapeutics can be designed entirely in a computer, much like how space missions or semiconductors are designed today,” he says. Kohl acknowledges the growing complexity of biological systems and the need for equally sophisticated safety frameworks as these powerful generative tools become more widespread. “Biology remains fundamentally messy,” he says. “A.I. currently amplifies our capabilities, but it still requires deep scientific intuition to ask the right questions and interpret what the models tell us.”

37. Timnit Gebru

• Founder & Executive Director, The Distributed AI Research Institute (DAIR)

As founder of the Distributed Artificial Intelligence Research Institute (DAIR), Timnit Gebru tirelessly advocates to keep ethics at the forefront of the A.I. conversation. Gebru launched DAIR, a community-based A.I. research institute “free from big tech’s pervasive influence,” in 2021 with $3.7 million from notable backers like the Ford Foundation, MacArthur Foundation and Open Society Foundation. 

Gebru is a vocal critic of artificial general intelligence, arguing its guiding framework is rooted in 20th-century Anglo-American eugenics. She contends that many of the same discriminatory attitudes—racism, xenophobia, classism, ableism and sexism—persist in today’s AGI movement, leading to systems that harm marginalized groups and centralize power while using the language of “safety” and “benefiting humanity” to avoid accountability.

She also examines A.I.’s role in geopolitics. In The New York Times last December, she argued that much technological progress is tied to government and military funding, not community needs. In an April essay for Scientific American, she warned that replacing federal workers with chatbots would be “a dystopian nightmare,” citing generative A.I.’s propensity to hallucinate even basic tasks like transcription.

Her ongoing influence on A.I. ethics has earned her accolades such as the 2025 Miles Conrad Award, a lifetime achievement honor from the National Information Standards Organization. Gebru is currently working on a memoir and manifesto, The View from Somewhere, scheduled for publication in fall 2026.

41. François Chollet

• Co-Founder, Ndea

François Chollet, one of A.I.’s most respected researchers and skeptics, has launched a new chapter in his career with Ndea, a research lab dedicated to advancing artificial general intelligence (AGI). Co-founded in January with Zapier co-founder Mike Knoop, Ndea blends frontier research with lessons from large-scale software engineering, aiming to rethink how AGI should be pursued. “The most pervasive and wrongheaded assumption today is that we can achieve human-level general intelligence by simply scaling up our current deep learning models, particularly large language models,” Chollet tells Observer. “My approach is to be very transparent about the limitations of our current technology and what we need to go beyond them. I’m not trying to sell a dream of imminent AGI.”

Chollet envisions A.I. making scientific discoveries the way humans do, and sees Ndea as the vehicle to unlock this potential. “The most brilliant minds are not attracted to turning a crank on a bigger machine. They are attracted to deep, fundamental scientific questions,” he says. “This attracts a different kind of talent, and that’s what we’ve seen at Ndea so far.”

Chollet is best known as the creator of Keras, the widely adopted deep learning library used at companies from YouTube to Spotify, as well as ARC-AGI, a benchmark regarded as one of the most important tests of genuine general intelligence in machines. His critiques of hype in the field, including his stance that current models fall short of true reasoning, have positioned him as both a visionary and contrarian voice in A.I. debates. “I’m worried about the effect that generative A.I. is having on human culture and communications. It is massively polluting our information ecosystem with synthetic content. This systematically devalues genuine human knowledge and expression,” he says. “The more we rely on generative A.I. for cultural creation, the more we will tend towards a stagnant state of perpetual remixing. Slop remixed from slop remixed from slop. Leaving us to sift through a digital landfill for a single original idea.”

Chollet and Knoop are also co-founders of the ARC Prize, a nonprofit focused on AGI development. Chollet authored the influential paper On the Measure of Intelligence.

43. Joy Buolamwini

• A.I. Researcher, Founder & President, Algorithmic Justice League

Joy Buolamwini, a Rhodes scholar, researcher and author of the 2023 book Unmasking AI: My Mission to Protect What Is Human in a World of Machines, has emerged as one of the most visible voices calling out bias in A.I. Her work blends advocacy, scholarship and public engagement, spotlighting the dangers of deploying A.I. systems that perpetuate racial and gender inequities.

In 2016, Buolamwini founded the Algorithmic Justice League, a nonprofit that combines art and research “to illuminate the social implications” of A.I. The group’s July 2025 report, The Comply to Fly, scrutinized the Transportation Security Administration’s facial recognition program in U.S. airports. After two years of study, it found troubling gaps in transparency around traveler rights and data handling, raising questions about a program initially described as “voluntary” but often operating otherwise. The report urged TSA to halt the rollout, while Buolamwini has been outspoken about how travelers can opt out of facial recognition at security checkpoints.

This year, she joined the inaugural Accelerator Fellowship Program at Oxford University’s Institute for Ethics in A.I., where she is focusing her research on addressing bias in algorithmic systems. Her scholarship continues to influence the field, including co-authored work with Timnit Gebru on disparities in algorithmic gender classification. Speaking at Duke University in February, Buolamwini framed the issue in cultural terms, asking her audience, “Show of hands. How many have heard of the male gaze? The White gaze? The postcolonial gaze? To that lexicon, I add the coded gaze, and it’s really a reflection of power. Who has the power to shape the priorities, the preferences—and also at times, maybe not intentionally—the prejudices that are embedded into technology?”

45. Jeff Dean

• Chief Scientist, Google DeepMind & Google Research

Under Jeff Dean’s leadership as Google’s chief scientist, the company has made major efficiency gains. In August, Dean published a paper showing that the carbon footprint of an average Gemini prompt fell 44x over the 12 months ending in May. He credits the progress to Google’s full-stack approach—more efficient model architecture and algorithms, custom hardware, optimized idling and ultra-efficient data centers. In 2024 alone, Google cut data center emissions by 12 percent and replenished 4.5 billion gallons of water, advancing its goal to return 125 percent of the freshwater it consumes across operations.

In May, Google launched Gemini AlphaEvolve, an A.I. system that autonomously generates and refines algorithms by combining creative models with testing systems. The tool has already boosted efficiency in data centers and A.I. training.

Dean’s influence extends beyond Google. Over the past two years, he has invested in 37 early-stage A.I. startups, including Perplexity, DatologyAI, Latent Labs, World Labs and Roboflow—often before Series A. In June, he backed the launch of the Laude Institute, led by Andy Konwinski, to translate academic breakthroughs into open-source tools, real-world products and socially responsible ventures.

He is also vocal about A.I.’s impact on work and research. At AI Ascent 2025, Dean predicted that advanced systems will soon reach the level of a junior software engineer, reshaping software development. At the Gemini Singapore Symposium earlier this month, he outlined how A.I. adoption could expand economic elasticity and spur job creation in tech-driven fields. He also forecast that A.I.-driven autonomous research could accelerate discovery and lower barriers to chip design, leading to an “explosion of specialized hardware.

47. Jahmy Hindman

• SVP & CTO, John Deere

At John Deere, Jahmy Hindman oversees the integration of A.I. into the agricultural pioneer’s industrial equipment, updating the tractors, combines and tillage machinery that generations of farmers have relied upon with automated, precision-guided features. Predictive maintenance engines, digital twins, and advanced analytics effectively turn each piece of equipment into a self-operating intelligence platform that leverages farming data to identify opportunities and increase crop yields. John Deere’s See & Spray technology, which detects individual weeds and applies herbicide only where needed, exemplifies this precision approach and has reduced herbicide use by up to two-thirds (saving farmers money). The company showcased its A.I.-powered fleet’s ability to handle new crops and conditions at CES 2025.

“With global food demand expected to rise as the population nears 10 billion by 2050, the need for efficiency and sustainability in agriculture has never been greater,” Hindman tells Observer. “Our customers operate in predominantly rural environments with changing and often harsh weather conditions. This is the place our technology must perform.”

Automation can fill in labor gaps across industries like agriculture, construction and commercial landscaping, according to Hindman. As John Deere goes full steam ahead on emerging technologies, the company is also partnering up with OpenAI to connect with prospective employees skilled in A.I.

Hindman is driving John Deere’s 2026 initiative to connect 1.5 million agricultural machines through satellite, enabling real-time remote control in regions lacking cellular coverage. The company has already achieved breakthrough connectivity speeds over satellite, which Hindman says “accelerates the model training flywheel and leads to faster, more robust improvements.”

In July, Hindman took on the role of a board member for Emergie Prairie, an innovation nonprofit. Hindman emphasizes the unique pressures of agricultural A.I. and the deep responsibility he feels to develop technology that farmers trust. The average farmer is 58 years old and works 12 to 18-hour days, and John Deere’s role is to “make every seed count, every drop count, and every bushel count,” Hindman says. “Farmers get one chance a year to do it right, so every decision and every action matters.”

49. Evan Solomon

• Minister of Artificial Intelligence, Digital Innovation and Federal Economic Development Agency for Southern Ontario

Evan Solomon’s appointment as Canada’s first-ever Minister of Artificial Intelligence in May represents a watershed moment in global A.I. governance. His mandate, rooted in the Liberal platform released in April, positions Solomon to influence A.I. policy through every aspect of Canada’s economy while addressing critical national security implications. With Prime Minister Mark Carney advocating for sweeping A.I. adoption to create an “economy of the future,” Solomon stands at the center of Canada’s most ambitious technological transformation effort. His role encompasses everything from incentivising business A.I. adoption through tax credits for small and medium enterprises to slashing repetitive government tasks and establishing an office of digital transformation.

Solomon faces the paradox of leading a nation with world-class A.I. research capabilities but leading citizen and business adoption. Despite Canada ranking number one globally in Academia-Industry Model Production Concentration and second in Foundation Models, only 12 percent of Canadian businesses reported using A.I. as of June, while 79 percent of Canadians express concerns about adverse A.I. outcomes. Solomon’s response has been both strategic and practical, as demonstrated by his recent $9 million announcement to fund the training of 5,000 mid-career energy sector workers in A.I. skills. His four-pillar approach—scale, adoption, trust and sovereignty—directly addresses these adoption barriers while leveraging Canada’s existing strengths, including the University of Toronto’s foundational work on LLMs and homegrown successes like Cohere.

The timing and potential impact of Solomon’s leadership make him particularly compelling to watch. Canada’s tech talent growth (66,000 jobs at a 5.9 percent growth rate) is outpacing the United States, with Toronto ranking third globally for tech talent behind only San Francisco and Seattle. Solomon’s efforts to retain Canadian talent, build data centers and reintroduce critical A.I. legislation (including the Artificial Intelligence and Data Act) could unlock enormous economic value, with some projections suggesting A.I. adoption could boost labor productivity by 17 percent over 20 years and generate $185 billion by 2030. As Canada ranks second in A.I. research output per capita among G7 nations yet 14th overall in Stanford’s global A.I. rankings, Solomon’s success in bridging this research-to-implementation gap could fundamentally reshape both Canada’s economic trajectory and global A.I. leadership dynamics. 

51. Daniel Gross

• Meta Superintelligence Labs

Daniel Gross, founder of venture firm NFDG, is guiding A.I. development as a member of Meta’s new superintelligence unit. Meta Superintelligence Labs (MSL) poached Gross and Nat Friedman, his investment partner, this summer, and Gross was hired to focus on developing consumer A.I. products within the newly formed division. 

Gross previously served as the CEO of Safe Superintelligence Inc. (SSI), the secretive A.I. startup launched last year by star researcher Ilya Sutskever, who now helms the company. The venture is focused on steering AGI research with safety at its core, a clear departure from Silicon Valley’s “move fast and break things” ethos. SSI secured $1  billion in initial funding in June  2024 and reached a $32  billion valuation by March 2025—without releasing a product. “The company’s future is very bright, and I expect miracles to follow,” said Gross in July.

The tech investor has a storied history of nurturing the launch of startups. As an angel investor, Gross has backed players like Character.ai, Weights & Biases, and Rippling. He was also formerly a partner at Y Combinator and helped launch its A.I. program and, alongside Friedman, set up the Andromeda Cluster—a supercomputer of over 4,000 GPUs that startups in their portfolio can rent—in 2023. With global-scale investment impact, AGI alignment, and ecosystem building, Gross is a multi-sector bridge‑builder shaping the next era of A.I.

53. Daphne Koller

• Founder & CEO, Insitro

As founder and CEO of Insitro, Daphne Koller has built strategic relationships with leading pharmaceutical companies: Eli Lilly for metabolic disease therapies, including metabolic-associated steatotic liver disease (October 2024), Bristol Myers Squibb providing $25 million for ALS genetic target research (December 2024), and Moorfields Eye Hospital for A.I. foundation models targeting neurodegenerative eye diseases (March 2025). These partnerships demonstrate that major industry players validate Insitro’s technology.

“My concern about A.I. in science isn’t the distant risk of superintelligence, but the erosion of rigor from the seductive plausibility of generative A.I.,” Koller tells Observer. “My core fear is that the ease of generating plausible answers will tempt organizations to bypass the hard-won ground truth of prospective, experimental validation. The danger isn’t that AI becomes too intelligent, but that we become complacent, trusting articulate outputs over real data. That would silently erode the scientific method and waste years chasing beautifully worded mistakes.”

Koller’s company operates with substantial resources despite recent strategic restructuring. Three years after founding, Insitro raised $400 million in 2021 for machine-learning-powered drug discovery. In May, the company cut 22 percent of staff (65 employees) to streamline operations and extend its runway into 2027, positioning itself for sustained development rather than expansion. This strategic focus on longevity reflects Koller’s understanding of biotechnology development timelines and capital efficiency requirements.

As co-founder and former co-CEO of Coursera, Koller democratized education globally before transitioning to A.I.-powered drug discovery. “Education taught me that A.I. shines when feedback loops are fast and data are abundant,” she says. “The biggest misconception is that traditional drug discovery is ready for A.I. You can’t just drop A.I. onto hundreds of incoherent spreadsheets and expect breakthroughs. We need to re-architect the systems and data collection around A.I. Done right, A.I. is an amplifier of rigorous biology, not a substitute for it.”

55. Stephen Schwarzman

• CEO, Blackstone

Under Schwarzman’s leadership, Blackstone has committed more than $100 billion to data centers, solidifying its position as the world’s largest private investor in A.I. infrastructure. These facilities, essential for training and deploying advanced models, are becoming as critical to A.I.’s growth as breakthroughs in algorithms and research. Schwarzman has guided Blackstone to focus not only on scale but also on strategic partnerships, ensuring that these data centers support both enterprise adoption and cutting-edge innovation across industries. This week, Blackstone announced it had agreed to buy a natural gas plant in Western Pennsylvania for nearly $1 billion, furthering its bet on the rising electricity demand of A.I. 

Beyond capital allocation, Schwarzman has shaped the broader discourse around A.I. through philanthropy and academic initiatives. His landmark gift to establish the MIT Schwarzman College of Computing placed ethics, governance and societal impact at the center of A.I. education, training a generation of leaders to navigate the promises and perils of machine intelligence. Through this combination of investment and thought leadership, Schwarzman wields influence over both the infrastructure powering A.I. and the frameworks guiding its responsible deployment.

57. Mati Staniszewski

Mati Staniszewski has a firm grip on A.I.-generated audio through his company ElevenLabs, a text-to-speech platform with customizable synthetic voices and voice cloning capabilities. A BlackRock and Palantir alumnus, Staniszewski has built ElevenLabs into a Series C startup worth a reported $6.6 billion that’s backed by big names like a16z and ICONIQ. In June, the company launched Eleven v3, supporting over 70 languages that can sigh, whisper, laugh and react. To date, ElevenLabs customers have created over 2 million agents through the platform, which have handled over 33 million conversations so far this year. In August, the company introduced Eleven Music, an A.I. music generation service. In September, ElevenLabs reached $200 million in annual recurring revenue.

“We believe that voice will be the primary interface for interacting with A.I. and technology,” Staniszewski tells Observer. “We are building a future where you can create conversational agents that help you speak to your tools, your content and your device—and they speak back, all in ways that feel natural. This changes how we create, access information and do business, while offering businesses a new way to express their brand identity.”

Staniszewski is outspoken about his company being supported by a slim team, doubling down on the move-fast startup mentality. That outlook has proven successful for ElevenLabs, which has partnerships with 60 percent of Fortune 500 companies, including Epic Games, Cisco and Chess.com (which offers an A.I. chess coach using their technology). A practitioner in licensing intellectual property for A.I. use, Staniszewski has worked with at least 5,000 creators, paid out around $5 million to use their vocal likenesses and made strides in the A.I. audio realm, all the while. “The future of A.I. hinges on trust,” he says. “And this is just the beginning.”

In September, Staniszewski told Bloomberg that ElevenLabs hopes to reach $300 million in recurring revenue by the end of this year.

59. Dr. Rumman Chowdhury

• Founder & CEO, Humaine Intelligence

Dr. Rumman Chowdhury advocates grounding A.I. in local realities and preserving human ingenuity at the center of its value. In 2022, she founded Humane Intelligence, a nonprofit focused on “bias bounties,” a concept she pioneered as Twitter’s director of machine learning ethics, transparency and accountability. The organization facilitates “institutionalized red teaming,” assessing A.I. systems for vulnerabilities, limitations and sociotechnical risks. Since 2023, Humane Intelligence has hosted more than 15 red teaming events across industry, government and academia worldwide. In Singapore, for example, testers from nine countries uncovered failures invisible in monolingual, monocultural lab settings. “Novel ideas originate in human minds,” Chowdhury tells Observer, urging organizations to “ensure decisions requiring creativity, ethical reasoning or contextual understanding are retained for humans” rather than delegated to A.I.

In 2024, Chowdhury joined New York City’s AI Steering Committee, created to oversee municipal adoption, foster public dialogue and guide agencies in deploying A.I. effectively. Alongside experts from IBM, Dell, Microsoft, Intel, Columbia University, NYU and more, she helps shape how algorithmic systems affect 8.4 million residents, from public benefits to policing.

“Cities face unique challenges,” she says. “Their problems are intensely practical, close to daily life and directly impact millions. Unlike federal regulators, city officials can’t simply issue broad principles. They must translate A.I. ethics into operational guidelines.”

Chowdhury has also advised at the federal level, serving on the Department of Homeland Security’s Artificial Intelligence Safety and Security Board and as the U.S. State Department’s Science Envoy for Artificial Intelligence. What concerns her most is that A.I. regulations are often treated as an afterthought. “If evaluations remain just a checkbox for compliance, rather than a meaningful process for stress-testing and improvement, we’ll end up deploying A.I. that’s brittle, unaccountable and out of step with people’s needs.”

61. Albert Bourla

• CEO, Pfizer

Under Albert Bourla’s leadership, Pfizer has committed to saving $5.7 billion in operational expenses by the end of 2027 through adopting a variety of A.I. tools. The pharmaceutical giant is on track to deliver $4.5 billion in net cost savings by the end of 2025, of which $500 million will be reinvested into R&D through 2026. 

In April, Pfizer inked its sixth deal with Flagship Engineering, a life sciences VC fund, to discover autoimmune drug candidates using A.I. Two months later, Pfizer expanded its collaboration with Chinese tech firm XtalPi to develop A.I. that can discover small molecules and predict crystal structures—building on their 2018 partnership that has already demonstrated A.I.’s ability to accelerate drug discovery timelines. Bourla’s A.I. investments position Pfizer alongside competitors like AstraZeneca, which have already reported significant reductions in drug discovery timelines thanks to A.I.

In the public sphere, Bourla has spoken about A.I.’s role in healthcare at the World Economic Forum this year. He regularly authors social media posts exploring A.I.’s potential to transform cancer care and close gaps in medical access. Bourla has articulated a bold vision for A.I. in oncology, predicting that within the next decade, A.I.-accelerated innovations like antibody-drug conjugates “could replace traditional chemotherapy, transforming cancer care and improving outcomes for millions.” 

63. Andrej Karpathy

• Founder & CEO, Eureka Labs

As founder and CEO of Eureka Labs,  launched in July 2024, Andrej Karpathy is building an A.I.-native education platform that pairs teachers with A.I. assistants to deliver scalable, personalized learning. The company’s first A.I.-powered course is already in development. Beyond Eureka, Karpathy reaches nearly a million subscribers on YouTube, where his tutorials on coding and large language models have become a global resource for aspiring A.I. practitioners.

Karpathy’s influence also extends into venture investing, where he has strategically backed companies with custom model development, agent platforms and core infrastructure. In May 2024, he joined the $25 million series A round for Lamini, an enterprise platform for custom LLMs founded by Google Cloud alum Sharon Zhou. He has since invested in /dev/agents, an A.I. agent development platform now valued at $500 million and Lambda, a Nvidia-backed infrastructure startup that raised $480 million in series D funding. 

His technical credibility stems from leadership roles as co-founder and former research scientist at OpenAI and former head of A.I. at Tesla, which provided a deep understanding of cutting-edge A.I. research and real-world deployment challenges. Karpathy has positioned himself as a key architect bridging A.I. research, education and venture capital influence across the ecosystem by combining educational innovation through Eureka Labs, mass learning outreach via YouTube and strategic investments in high-value startups.

65. Lila Ibrahim

• COO, Google DeepMind

Lila Ibrahim is one of the most influential executives shaping the path toward artificial general intelligence (AGI). As chief operating officer of Google DeepMind, she oversees operations, infrastructure and risk governance at one of the world’s most important A.I. research labs. Her mandate is to ensure that the lab’s cutting-edge science is translated into products and systems that benefit humanity at scale.

Ibrahim brings experience from senior roles at Intel and Coursera, giving her a unique vantage point at the intersection of hardware, software and education. Under Ibrahim’s leadership, DeepMind’s breakthroughs power core Google products used by billions. Gemini, Google’s family of generative A.I. models, is now embedded across Search, Workspace, and Cloud, reshaping how users interact with information and productivity tools. She has also been central in building DeepMind’s governance framework, balancing the company’s ambition to develop AGI with its responsibility to mitigate risks.

Beyond her executive role, Ibrahim is an influential voice in the global A.I. policy conversation. She has publicly advocated for thoughtful, balanced regulation and international coordination, cautioning that overly restrictive policies could stifle innovation while failing to address real risks. 

Her contributions have been recognized worldwide. Ibrahim has been honored by the Financial Times, the United Nations and the World Economic Forum, underscoring her role as both an industry leader and a global policy influencer.

68. Percy Liang

• Associate Professor, Stanford University & co-founder, Together AI

Percy Liang is a senior fellow at Stanford’s Human-Centered Artificial Intelligence (HAI), a program co-led by Fei-Fei Li. As director of the Center for Research on Foundation Models (CRFM), Liang develops evaluation frameworks for leading foundation models and transparency grading systems for A.I. model companies. With more than 180,000 Google Scholar citations across 58 publications, he is a prominent researcher laying the groundwork for A.I. innovation. His recent studies—ranging from robot policies to large language models for medical tasks—shape the integration of A.I. into daily life and the broader pursuit of artificial general intelligence (AGI).

Liang takes a contrarian stance on AGI, one of the field’s most widely discussed concepts. “The notion of AGI as a goal is misguided (at best it’s a vibe),” Liang tells Observer. He argues that A.I. already outperforms humans in many areas while lagging in others, and will continue improving unevenly across different dimensions “without any natural bounds.”

Liang’s biggest concern is what he sees as a troubling decline in openness. “I worry deeply about the decline of openness in A.I.,” he says. While models like Llama and DeepSeek are technically open-weight, he argues they fall “far from the true spirit captured by open-source software.” In his view, the current ecosystem fosters “a few producers and mostly consumers,” in stark contrast to the Internet revolution’s decentralized, open-source foundations. “Today, we need the same for A.I.,” Liang emphasizes.

The DeepSeek breakthrough marked a pivotal moment in his view of the A.I. landscape. “It was celebrated by proponents of open(-weight) models, but it was also notable coming from a non-U.S. (Chinese) company,” he says. The event underscored both the global nature of A.I. progress and the ongoing debate over model accessibility.

As an educator, Liang advocates a dual approach to teaching with A.I. He believes students must first demonstrate “foundational mastery” by completing basic tasks without A.I., while also learning to embrace it as a tool that can extend their capabilities. This balance, he argues, maximizes A.I.’s educational potential while addressing concerns about surveillance.

Liang co-led this year’s Stanford AI Index Report, an influential HAI publication for policymakers and industry leaders. His goal is to make foundation models more accessible and transparent by advancing open-source technology and rigorous benchmarking.

70. Walid Mehanna

• Chief Data & A.I. Officer and Chairman of the Digital Ethics Advisory Board, Merck KGaA, Darmstadt, Germany

As Chief Data & A.I. Officer and Chairman of the Digital Ethics Advisory Board for Merck KGaA, Darmstadt, Germany, Walid Mehanna has overseen one of the most comprehensive enterprise A.I. deployments in pharma, life sciences and electronics (per the Pharma A.I. Readiness Index). Under his leadership, Merck KGaA rolled out its optimized myGPT suite application via a 2023 partnership with A.I. startup Langdock. Today, 52,000 of the company’s 63,000 employees use the suite, with 14,000 digital assistants active across 1,600 groups and more than 12 million prompts logged—recently approaching 1.5 million per month. A company representative says it is the most-used non-mandatory tool.

“The culture of a science and technology company helps A.I. adoption immensely,” Mehanna tells Observer. “Scientists are used to the ‘experiment and iterate’ phase of any methodology. That’s why it’s a great place to pioneer A.I.” Mehanna compares his method for building a corporate A.I. strategy to scaling a pyramid: productivity for all (myGPT) is the foundation; operational A.I. in core business streams is next; and advanced A.I. for products and revenue sits at the top. “Models will change, but our data, guardrails and talent compound, so the strategy doesn’t,” he explains.

Mehanna joined Merck KGaA from Mercedes-Benz, where he was the head of data and analytics. “In a car, you can instrument almost everything. In biology and chemistry, you rarely see the full system,” Mehanna says. “That lack of visibility turns modeling, validation and guardrails into the real engineering challenge. Data is noisier, ground truth is murkier, and transparency is limited. So A.I. needs to be more humble, explainable, and evidence-driven.”

Since 2021, Mehanna has convened digital ethics experts from Mayo Clinic, Johns Hopkins University and Dana-Farber Cancer Institute to implement ethical frameworks for Merck KGaA’s A.I. deployment across the whole value chain and business sectors. Mehanna doesn’t believe “that you must choose between speed and responsibility. With the right guardrails, you move faster because you can innovate with confidence. Governance is like brakes: the better they are, the faster you can drive safely.”

72. Sara Hooker

• Frontier A.I. Leader

As the former vice president of research and head of Cohere Labs at Cohere, Sara Hooker led the $6.8 billion company’s nonprofit research arm. During her three years with Cohere, Hooker’s team of fewer than 30 engineers and researchers published over 100 research papers, collaborated with over 150 institutions and released models that have been downloaded 23 million times. She left Cohere in September 2025 to explore “new problems” that are “central to the future of intelligence,” she said when announcing her departure.

While tight-lipped about her next venture, Hooker confirmed to Observer that she’ll begin her new role in early October. “Humans adapt,” Hooker tells Observer. “Our current A.I. doesn’t. The biggest leap that is going to come in building thinking machines is solving for this gap.”

Hooker’s focus on multilingual language models ensured A.I. accessibility across more than 100 languages worldwide, driven by her conviction that “when you speak in someone’s language, you really connect with their heart, not their head.”

Hooker’s research has exposed critical flaws in A.I. evaluation and transparency. In April, she led a study revealing that LM Arena, maker of the influential Chatbot Arena benchmark, systematically biased results to favor certain companies—including OpenAI, Meta, Google and Amazon—at the expense of their rivals. This research challenges the integrity of widely used A.I. evaluation systems and demonstrates her role in holding the industry accountable for fair assessment practices.

Through Cohere Labs, Hooker advanced open science collaboration in machine learning. Projects like Aya focused on multilingual model development and research into model efficiency and data quality. By leading international research initiatives, challenging industry evaluation standards, and expanding multilingual A.I. accessibility, Hooker has emerged as a key architect of more equitable and transparent A.I. development.

74. Chetan Dube

• CEO & Chairman, Quant

After selling his conversational A.I. company Amelia to SoundHound for $180 million in 2024, Chetan Dube launched Quant, a company focused on agentic A.I. for customer service that is reimagining labor, accountability and A.I.’s role in the economy. In less than a year, Quant has secured enterprise deals with a global airline, a major utility company and a telecommunications consortium. One of the largest utilities in the country is now resolving over 76.8 percent of all its inbound complex calls through Quant’s digital employees, Dube tells Observer.

A former NYU assistant professor and longtime A.I. entrepreneur, Dube has spent nearly three decades working to replicate human intelligence in software, collaborating with governments in the U.S., Europe and the Middle East to adopt A.I. across their tech stacks. Accruing a net worth of $2.4 billion, Dube has been named one of Forbes’ top 10 minds in A.I. He puts trust and safety at the core of his agenda, calling for a new safety governance model where A.I. systems are dissected, tested and certified before deployment. 

He also advocates taxing digital employees and redistributing the gains to human workers. Dube argues that while traditional A.I. delivered “marginal” single-digit ROI, agentic technologies can achieve returns exceeding 50 percent by focusing on action rather than information. “A.I. companies are often advancing generative technologies for information retrieval, which accounts for 19 percent of the total incoming volume, as opposed to Agentic technologies for actions that handle 81 percent of incoming tasks,” he says.

Dube tells Observer that the urgency driving his work became personal when his son asked him whether he was going to be a “robodad.” That moment crystallized his concern that “most people aren’t taking seriously” the possibility of A.I.’s more dystopian outcomes. “Humans aren’t ready for a hybrid society,” where we can’t distinguish between digital and human colleagues, which Dube believes will be here by 2030. Humanity must prepare for what he calls a “digital tsunami” that could either clean up humanity’s problems or lead to social unrest. “Man must proactively move to a higher ground of creative thinking,” Dube says. “We have a revolutionary power at hand. If we harness it gainfully, it can cure the planet of many maladies; if we don’t, it can be the final invention.”

77. Hiroaki Kitano ' to the clipboard">

• President & CEO, Sony Computer Science Laboratories, Inc.

As president and CEO of Sony Computer Science Laboratories, Hiroaki Kitano oversees A.I. strategy in entertainment, robotics and consumer technology, reaching hundreds of millions of users worldwide. He also directs Sony Global Education and founded the RoboCup Federation, which advances robotics and A.I. through international competitions and social impact projects.

In June, Kitano was appointed Senior Executive Advisor for Niremia Collective, a Silicon Valley venture capital firm specializing in well-being technology. The role positions him to channel significant investment into A.I. applications for healthcare and wellness sectors. At the World Economic Forum in January, he articulated his vision that A.I.’s impact has “propelled science at unprecedented speed, which will change the form of civilization for decades to come,” emphasizing the need for cross-disciplinary A.I. development rather than siloed approaches.

Kitano frames A.I. as the fourth great civilizational shift, “the early stage” of an A.I. Industrial Revolution. “We may still be at the ‘steam engine’ phase,” Kitano tells Observer, noting the qualitative differences between today’s revolution and the past lies in autonomous A.I. “This will amplify the capabilities of our civilization,” he says.

Kitano’s ultimate goal is to develop “A.I. as a scientist,” capable of making Nobel Prize-worthy discoveries, through his proposed Nobel Turing Challenge. “There are human cognitive limitations,” he tells Observer. “A range of issues hampers human scientists from understanding the mechanisms behind very large, complex, dynamic systems. A.I. for autonomous scientific discovery can overcome such limitations of human cognition to perform the kind of science a human cannot.”

​​Beyond corporate strategy, Kitano actively advances research equity through concrete initiatives. In February, he orchestrated the inaugural “Sony Women in Technology Award with Nature,” distributing $250,000 grants to three women researchers to accelerate their technological research. Through RoboCup Federation and his venture capital advisory role, Kitano leverages Sony’s broad technology portfolio to position himself as a key architect at the intersection of A.I. innovation, investment strategy, and inclusive research development worldwide.

79. Nitzan Mekel-Bobrov

• Chief A.I. Officer, eBay

As eBay’s chief A.I. officer since 2021, Nitzan Mekel-Bobrov has integrated generative A.I. across a marketplace serving millions of users worldwide. He led the development of eBay’s “magical listing tool,” which allows sellers to upload product images while A.I. automatically extrapolates data and fills in key details—a feature with a success rate exceeding 90 percent.

In May, eBay launched image-to-video capabilities to enhance the selling experience, along with its first consumer-facing shopping agent designed to help buyers discover “hidden gems.” In August, the company rolled out “Offers in Messaging,” which uses generative A.I. to pre-write responses to prospective buyers, and an automated feedback tool that leaves positive reviews on seller pages when products are delivered without issues and the buyer opts not to leave a review.

Under Mekel-Bobrov’s leadership,  eBay has increased investments in GPU computing to support more advanced A.I. functionality across the platform. The company’s stock price reached an all-time high in August, prompting Fortune to ask, “Is eBay actually sexy again?”

Last week, eBay CEO Jamie Iannone told The Wall Street Journal that the company aims to become an A.I. leader, leveraging thirty years of data from billions of transactions as it invests in generative technology. In the same article, Mekel-Bobrov noted that all 11,500 eBay employees now use A.I. agents for project-based work, tracking every task and reducing meeting volume by double digits. Each employee has the resources to build personalized A.I. agents tailored to their job function and specific needs.

81. Andrew Ng

• Founder, DeepLearning.AI

Andrew Ng, an A.I. pioneer and evangelist, is moving the needle by countering one of Silicon Valley’s biggest myths: namely, that A.I. is an all-purpose magic wand. The founder of DeepLearning.AI (an ed-tech platform focused on artificial intelligence) and AI Aspire (an A.I. advisory firm), and managing general partner of the AI Fund (a venture capital firm), has recently asserted that the real reason most startups struggle is poor execution, not bad code. And his thesis—that as tech is optimized, humans need to change to keep up—is reshaping how founders think about building, scaling and deploying A.I. 

His eyes-wide-open, science-first outlook is tempered with optimism—at this year’s GCVI Summit, he was quick to point out that the hype surrounding A.I. shouldn’t put off investors. More importantly, he practices what he preaches. DeepLearning.ai has now equipped more than 7 million learners worldwide with the knowledge to deploy A.I. in the real world. Under his leadership, the AI Fund expanded globally in late 2024; this year, it hosted a Buildathon, during which developers shipped full apps in mere hours using A.I.-assisted coding. Recently, Bain & Company tapped Ng and AI Aspire for a partnership that will bring his expertise to clients looking for sharper artificial intelligence strategies.
to
What makes Ng so vital is his ability to democratize a deeply complex field. Whether teaching (on top of everything else, he’s an adjunct at Stanford), investing in underrepresented markets or informing policy, he champions an inclusive A.I.-powered future where everyday builders—not just elite labs and VC-backed unicorns—are using the technology in meaningful ways to solve real-world problems across sectors and geographies.

83. Jakub Pachocki

• Chief Scientist, OpenAI

Promoted to chief scientist in 2024, Jakub Pachocki now leads OpenAI’s research agenda to build models capable of reasoning through complex scientific and mathematical problems. OpenAI CEO Sam Altman has described him as “one of the greatest minds of our generation.”

A veteran of the company for eight years, Pachocki helped shape many of OpenAI’s defining milestones, from leading the team whose bots defeated the world champions of Dota 2 in 2019 to training GPT-4. More recently, he has focused on advancing OpenAI’s “reasoning models,” including o1, which takes time to deliberate before producing answers to complex coding, math and scientific challenges. The results came quickly. In 2024, OpenAI’s reasoning models earned gold-medal performance at the International Mathematical Olympiad and outperformed nearly all human competitors in a major coding contest. 

Pachocki also sees broader scientific potential in A.I. In May, he told Nature that mounting evidence suggests systems can generate novel insights. “Even this year, I expect that A.I. will, maybe not solve major science problems, but produce valuable software, almost autonomously,” he said. His remarks helped spark debate about machine learning and its implications for research and industry.

Though less publicly visible than some peers, Pachocki’s work underpins models now deployed by hundreds of millions worldwide. He’s also one of the researchers challenging the notion that A.I. is an indecipherable “black box” system. “I emphasize seeking understanding of how deep learning works,” he told TIME recently. “Despite it seeming like it’s just mathematics, it’s really a sort of natural science, where you’re trying to understand this phenomenon.”

85. Chris Olah

• Co-Founder, Anthropic

Christopher Olah is at the forefront of efforts to make large language models more transparent, tackling one of A.I.’s most enduring mysteries: how neural networks actually make decisions. As a technical staff member and co-founder at Anthropic, Olah leads groundbreaking research in mechanistic interpretability, developing methods to peer inside the “black box” of A.I. systems and map their internal reasoning processes.

His work has already delivered major breakthroughs. Last May, his team identified groups of neurons tied to specific behaviors such as bias detection and spam filtering—and showed that deliberately altering these clusters could change a model’s behavior in predictable ways. This gave researchers a new lever for alignment and control. In July, his team extended its methodology to attention mechanisms, advancing the “attribution graph” technique for tracing how inputs flow through and shape outputs. This marked a shift from treating neural networks as opaque algorithms to viewing them as interpretable, steerable systems.

Olah’s work addresses one of A.I. safety’s most urgent challenges: understanding and controlling increasingly powerful systems before they grow too complex to govern. His research in mechanistic interpretability is laying the groundwork for more reliable, transparent, and safe A.I. deployment.

88. Abdullah bin Sharaf Alghamdi

• Founder & President, Saudi Data & A.I. Authority (SDAIA)

Professor Abdullah bin Sharaf Alghamdi has emerged as a key architect of Saudi Arabia’s A.I. transformation under the kingdom’s Vision 2030. As the founding leader of SDAIA and chair of the Saudi Federation for Cybersecurity, Programming and Drones (SAFCSP), Alghamdi has positioned the Kingdom at the forefront of national-scale A.I. governance and digital strategy.

Since launching SDAIA in 2019, Alghamdi has steered major initiatives in data infrastructure, A.I. ethics and digital economy development, ensuring Saudi Arabia is an adopter of A.I. and a global player in shaping its governance. Under his leadership, SDAIA has struck strategic partnerships with international tech firms and governments, most recently signing a memorandum of understanding with AMD at the Saudi-U.S. Investment Forum in June 2025 to bolster data center development. Earlier agreements with U.S. counterparts have also underscored his role as a bridge between Saudi ambitions and global A.I. ecosystems.

Alghamdi’s influence is institution-building. With SDAIA and SAFCSP, he established the infrastructure and talent pipelines underpinning the Kingdom’s A.I. strategy. His authority spans multiple sectors, from digital economy growth to cybersecurity, and his initiatives align with global debates on ethical A.I., inclusion and governance. As A.I.’s role in the global economy accelerates, Alghamdi’s leadership ensures Saudi Arabia continues to emerge as a significant contributor to the worldwide A.I. agenda.

90. Robert Opp

• Chief Digital Officer, United Nations Development Programme (UNDP)

As Chief Digital Officer of the United Nations’ Development Programme, Robert Opp shapes how A.I. and data drive economic growth across more than 170 countries. From digital public goods to A.I.-powered tools, he focuses on making emerging technologies accessible and inclusive, but challenges the assumption that A.I. delivers benefits equally everywhere.

“In reality, the benefits have been distributed unequally across and within countries,” Opp tells Observer, noting that A.I. amplifies exclusion when data sets and solutions don’t reflect local realities, languages or cultural context. “What’s missing in the conversation is how to localize A.I.,” he explains. “Without diverse and inclusive datasets, A.I. will continue to misrepresent and even marginalize entire populations. This issue doesn’t make headlines as much as job displacement or safety risks, but it is fundamental to whether A.I. can actually serve everyone.” 

Opp prioritizes building foundations—digital IDs, payment systems and data exchanges—before layering A.I. on top. He has become a leading voice in international development circles, advocating for A.I. equity as a driver of private-sector growth and the UN’s Sustainable Development Goals. In commentary on South Africa’s A.I. strategy, he underscored the chance for developing nations to “prioritize equity, inclusion and rights from the start, rather than retrofitting protections later.”

A turning point in his thinking came with an MIT report in August showing 95 percent of companies had seen “zero ROI” from generative A.I. “That felt like a real moment,” he says, “a signal we might finally be moving past the hype cycle.” The finding reinforced his focus on rigorous evaluation of A.I. investments, especially in the public sector.

A.I.’s “dizzying gallop” was also a theme of the 2025 UNDP Human Development Report, published in May. Under Opp’s leadership, UNDP is piloting A.I. for agriculture (real-time crop feedback), health (maternal care access through language models) and education (personalized, inclusive learning). “These are areas where A.I. directly improves lives,” Opp says. “But only if countries have the infrastructure, data and governance to make it work.” He credits UNDP’s “people-first approach” and initiatives like the AI Trust and Safety Re-imagination Programme with building data governance and privacy protections for vulnerable populations.

Before joining UNDP, Opp was Director of Innovation at the World Food Programme, where he launched ShareTheMeal, an app that raised $40 million to fight hunger. The experience taught him that “digital platforms can radically reduce the friction of engagement. When people can act instantly from their phones, they are more willing to participate, even in small ways.” Today, Opp brings that philosophy to embedding ethical A.I. design at the heart of humanitarian work.

92. Anusha Dandapani

• Chief Data & Analytics Officer, United Nations International Computing Center (UNICC)

As head of the United Nations International Computing Center’s new A.I. Hub, Anusha Dandapani bridges the gap between the public, academic and private sectors. A working adjunct professor at New York University and Fordham University, Dandapani’s coordination across the United Nations helps build cohesive strategies for responsible A.I. governance, development and deployment. 

“We’re in a ‘move fast and break things’ moment, but the guardrails for responsible use of A.I. are not yet fully in place,” Dandapanni tells Observer. “How do we encourage rapid innovation while ensuring protections that serve humanity at large? It’s a bit like when cars were first introduced: we put them on the road long before we had seatbelts and airbags. We can’t afford to repeat that mistake with A.I.”

Working within UN mandates, Dandapani has collaborated on tangible achievements: Last year, UNICC partnered with Columbia University to support the creation of the Gender Bias Overview Tool (GenBOT), which uses A.I. and machine learning to “identify, measure and address gender biases in datasets and technological solutions.” A few months later, Dandapani spearheaded UNICC’s collaboration with NYU to develop the A.I.-Driven Media Analysis Tool, which identifies “xenophobic language, misinformation and harmful narratives in media reporting” on migration and displacement issues.

Dandapanni introduced the UNICC A.I. Hub in July, at a launch event coinciding with the AI for Good Summit in Geneva. UNICC’s A.I. Hub operates under the organization’s strategic framework through 2030, offering consultancy, shared and custom solutions and training programs for UN teams through their A.I. Academy. The A.I. Hub’s global influence is poised to help UN member states effectively leverage A.I. as it evolves. 

94. Rohit Prasad

• SVP and Head Scientist of AI, Amazon

Rohit Prasad drives Amazon’s dual push into consumer voice technology and artificial general intelligence. As SVP and Head Scientist of AI, he oversees the company’s generative A.I. portfolio, including Amazon Nova, the foundation model designed for frontier intelligence with industry-leading price performance. In April, Amazon introduced Nova Sonic, a real-time, emotion-aware speech model that interprets tone and intent for more natural interactions. Prasad calls it a “huge step” toward AGI, advancing models that grasp context and complexity. Amazon says Nova Sonic is nearly 80 percent faster and cheaper than rival voice systems like OpenAI’s GPT-4o. Early adopters include ASAPP, Education First and Stats Perform, which deploy it across customer support, language learning and sports analytics.

Prasad’s shift into AGI leadership in 2023 built on his years as head scientist for Alexa, where he led multidisciplinary teams advancing far-field speech recognition, natural language understanding and the machine learning breakthroughs that made Echo’s hands-free interaction mainstream. Under his leadership, Alexa became a global standard for ambient intelligence—A.I. that blends seamlessly into environments and adapts across devices. Today, those two worlds converge. Amazon Nova foundation models now power Alexa+, handling over 70 percent of responses, while Nova Act drives Alexa’s Web Action SDK for voice-controlled web interactions. This integration ties Prasad’s track record in conversational A.I. directly to Amazon’s broader AGI ambitions.

Beyond voice tech, Prasad is advancing safety through initiatives like the Nova AI Challenge, which used penetration testing to stress-test model trustworthiness in code generation. He is a named author on more than 100 scientific papers, holds multiple patents, and serves on the board of the Partnership on AI, a nonprofit focused on ensuring artificial intelligence advances human-centered outcomes.

96. David Sacks

• Founder & Partner, Craft Ventures

A former PayPal executive and Trump-appointed tech advisor, David Sacks channels his influence as founder and partner of Craft Ventures, an early-stage venture capital firm. As the White House’s A.I. and crypto czar since 2025, Sacks received an ethics waiver allowing him to participate in policy decisions, despite relevant financial holdings and business relationships that have drawn scrutiny.

His role became controversial when Sacks advocated for a UAE chip deal that would grant the Emirates access to hundreds of thousands of advanced A.I. chips, despite administration colleagues expressing concerns about his longstanding business relationships with the Gulf and past investments in the A.I. industry. The timing raised questions as the Abu Dhabi Investment Authority, which oversees $1.5 trillion in sovereign wealth under Sheikh Tahnoon bin Zayed Al Nayhan, was an early investor in Craft Ventures. Just weeks before the chip deal approval, Al Nayhan’s firm invested $2 billion in World Liberty Financial, a Trump-linked cryptocurrency company.

Beyond his government role, Sacks continues his venture capital work through Craft Ventures. In 2025, he led major investments worth tens of millions into startups, including Vultron, Norm AI and Horizon—companies developing A.I. agents and safety infrastructure. His investment portfolio spans various A.I. applications, from automation tools to safety-focused technologies.

Sacks has also become an outspoken voice in A.I. policy debates, often framing issues through a competitive lens with China. In mid-2025, he warned that overreaching chip export restrictions could undermine U.S. competitiveness as Chinese models close the performance gap, arguing, “Do we want these countries to be piggy banks for American A.I. or for Chinese A.I.?” He has tempered fears of A.I.-driven job loss, instead advocating for attention to digital-era mental health, calling it a “tech psychosis” that demands structural reform.

98. Fidji Simo

• Applications CEO, OpenAI

Fidji Simo has outlined an ambitious vision for A.I. as a democratizing force. In July, ahead of beginning her new role at OpenAI, Simo distributed what Wired called a “hyper-optimistic” welcome memo to OpenAI staff, describing how ChatGPT could serve as a personal coach, tutor and emotional companion available to everyone—drawing from her own experience with a business coach she calls “transformative” but acknowledges was “a privilege reserved for a few.” 

Simo positions A.I. as capable of compressing “thousands of hours of learning into personalized insights” and helping users “develop confidence in areas that once felt opaque.” Her practical mandate involves translating OpenAI’s research into viable commercial products and securing the business partnerships to fund such expansive promises.

Simo formerly served as the CEO of Instacart. She’s credited for successfully taking the company public and launching initiatives like the A.I.-powered Caper smart cart and a healthcare research institute focused on food access and insecurity. She joined OpenAI’s board in 2024. Now, as an official member of the executive team, she is tasked with making the company profitable as it gears up for its own potential IPO. 

Since August, Simo has reported directly to Sam Altman, who calls her “uniquely qualified” to lead a group of business and operational teams “responsible for how our research reaches and benefits the world.” Earlier this month, Simo published a blog post on expanding economic opportunity with A.I., as the company announces its job matching platform and certification program under the banner of democratizing economic opportunities. While OpenAI executives acknowledge that A.I. is reshaping the job market and creating disruption, the company positions its new commercial platform as a pathway for workers to adapt to the changing landscape that A.I. technologies have helped accelerate.

100. Sam Hamilton

• Sustainable & Responsible A.I.

As head of data and A.I. for Visa, a role he stepped down from on September 12, Sam Hamilton was a key architect of the company’s global A.I. strategy, ensuring the technology delivered measurable business impact while reshaping how billions of people engage with digital payments. During his tenure, Hamilton oversaw more than $10 billion in investments across fraud prevention and advanced analytics. These efforts not only reinforced Visa’s leadership in secure payments but also produced tangible results, including blocking an estimated $40 billion in fraudulent activity in 2023.

“Most people focus on model performance or bias, but few talk about how dependent we are on opaque data pipelines, proprietary training sets and third-party infrastructure,” Hamilton tells Observer. “If a foundational model is trained on flawed or manipulated data, or its lineage is unclear, it can introduce systemic risks across industries—especially in finance.”

In March 2024, Hamilton spearheaded the launch of Visa Protect, a suite of A.I.-powered tools designed to stop fraud in real time, with a particular focus on card-not-present and digital transactions. The rollout reflected his vision for A.I. at scale: seamlessly embedded into financial infrastructure to enhance trust, efficiency and security. The following month, Visa debuted Visa Intelligent Commerce, a partner program with platforms including Anthropic, Mistral, OpenAI and Perplexity, allowing developers to leverage Visa’s A.I. commerce capabilities worldwide.

Hamilton’s next venture, announced soon, will focus on sustainable, responsible and ROI-backed A.I. solutions for individuals, enterprises and communities. Looking back, he points to the release of multimodal capabilities as a turning point. “When LLMs began interpreting images alongside text, it marked a major leap,” he says. “Suddenly, A.I. could understand visual context, diagrams, screenshots and even handwriting. This unlocked new use cases in education, accessibility, design and diagnostics—blurring the lines between human and machine perception.”

Reflecting on his time at Visa, Hamilton lauds the company’s commitment to being “accountable stewards of data,” citing the principles and safeguards that guided its A.I. systems. “Visa has been using A.I. for more than 30 years,” he says. “Visa invests in and employs A.I. to drive innovation and support its mission to uplift everyone, everywhere by being the best way to pay and be paid.”