Meta's Top AI Whiz Joins Controversial Startup—Is This the Next Big Tech Revolution or a Risky Gamble?

In a significant move within the competitive landscape of artificial intelligence, Soumith Chintala has joined Thinking Machines Lab, marking his first role since departing from Meta earlier this month. Chintala, known for co-creating PyTorch, a widely-used open-source AI framework, expressed enthusiasm for his new role on his X account, stating, "[T]hinking machines—the people are incredible."

Thinking Machines Lab, which was unveiled in February by Elon Murati after her exit from OpenAI, positions itself as an AI research and product lab focused on enhancing human-AI collaboration. In recent months, the startup has actively sought top-tier talent, recruiting individuals from notable companies including Meta, OpenAI, and Anthropic, as well as experts from academia.

Chintala's transition comes amid significant restructuring within Meta's AI divisions. The company has been overhauling its AI strategy, recently hiring a plethora of talent from its rivals, including Google DeepMind and Apple. This shift is reportedly part of the establishment of a new division called Superintelligence Labs, led by Alexandr Wang, the former CEO of Scale AI. Underlying these changes is the possible departure of Yann LeCun, Meta's chief AI scientist and founder of the Facebook AI Research (FAIR) group, which Chintala was integral in building.

Although Thinking Machines Lab declined to comment on Chintala's hiring, the startup has been aggressively competing in the "AI talent wars." Its early hires and advisors include prominent figures such as John Schulman, who co-led the development of ChatGPT, and Alec Radford among others. Reports indicate that the company is offering salaries as high as $500,000 for technical roles, with many staffers making between $450,000 and $500,000 before bonuses and equity.

Earlier this year, Thinking Machines successfully raised a monumental $2 billion seed round at a valuation of $10 billion. Now, the startup is in talks to secure additional funding, which could elevate its valuation to $50 billion, according to Bloomberg.

For Chintala, this transition represents a fresh chapter after his 11-year tenure at Meta and nearly eight years of leading the PyTorch project. Reflecting on his time at Meta, he noted that PyTorch had become a crucial tool in use by "virtually every major AI company" and was taught in classrooms from prestigious institutions like MIT to rural India. In his farewell message, he indicated readiness for new challenges, stating he was prepared to try "something small, something new, something uncomfortable."

Chintala also acknowledged the rapid evolution of AI, suggesting that PyTorch, while stable, might struggle to keep pace without his guidance. Meanwhile, Thinking Machines has not yet launched its full range of products, though its inaugural tool, Tinker, is already being used by researchers at Princeton and Stanford, along with early business customers.

Despite the excitement surrounding Thinking Machines Lab, it has faced its own challenges, including staff departures. Notably, co-founder and former Meta researcher Andrew Tulloch left the company in October to return to Meta, as reported by The Wall Street Journal. As the AI landscape continues to evolve, the implications of these movements in talent and resources will be closely watched by industry experts and competitors alike.