The AI Talent Wars Intensify: Meta Poaches OpenAI's Trapit Bansal for "Superintelligence" Push

The battle for top-tier artificial intelligence talent is heating up, and Meta Platforms is making aggressive moves. The tech giant has reportedly secured Trapit Bansal, a key AI researcher who significantly contributed to OpenAI's early work in AI reasoning and reinforcement learning, for its newly formed AI superintelligence team. This hire signals Meta's ambitious drive to compete at the forefront of AI development, directly challenging industry leaders like OpenAI and Google.

Bansal, who joined OpenAI in 2022, is now among the most prominent figures to depart the firm and join Meta's burgeoning AI superintelligence unit. This initiative, designed to attract the brightest minds in the field, is a clear sign of Meta's commitment to pushing the boundaries of AI. An OpenAI spokesperson, Kayla Wood, confirmed Bansal's departure, and his LinkedIn profile also indicates his move from OpenAI this June.

A Critical Loss for OpenAI's Reasoning Efforts

During his tenure at OpenAI, Bansal collaborated closely with co-founder Ilya Sutskever. He played an instrumental role in the development of "o1," the company's foundational AI reasoning model. This move is particularly impactful given the escalating interest in AI reasoning models, especially as rivals like OpenAI's "o3" and DeepSeek's "R1" continue to achieve new performance milestones.

Joining a Star-Studded AI Team at Meta

Bansal brings his specialized expertise to an already impressive roster within Meta's AI superintelligence lab. This powerhouse team includes industry luminaries such as former Scale AI CEO Alexandr Wang, ex-Google DeepMind researcher Jack Rae, and machine learning veteran Johan Schalkwyk. Furthermore, recent reports from Bloomberg and The Wall Street Journal indicate that several other former OpenAI researchers, including Lucas Beyer, Alexander Kolesnikov, and Xiaohua Zhai, have also recently joined Meta.

The core mission of this elite lab is to develop next-generation AI reasoning models that can match or even surpass the capabilities of those from OpenAI and Google. While Meta has not yet publicly released an AI reasoning model, the composition of this new team underscores its serious intent.

Mark Zuckerberg's Aggressive Talent Acquisition Strategy

Meta CEO Mark Zuckerberg is personally driving this AI hiring spree, reportedly offering compensation packages in the range of $100 million to lure top AI talent. While the specifics of Bansal's offer remain undisclosed, it is clear that Meta is sparing no expense to build its new AI unit.

Zuckerberg has also reportedly explored acquiring startups with formidable AI research labs, such as Sutskever's Safe Superintelligence, Mira Murati's Thinking Machines Labs, and Perplexity, to further bolster Meta's AI capabilities. However, these acquisition discussions have not yet materialized into final agreements.

OpenAI CEO Sam Altman, in a recent podcast, openly acknowledged Meta's attempts to poach his startup's top talent. Despite these efforts, he asserted that "none of our best people have decided to take him up on that," suggesting a resilience within OpenAI's core team. However, with the confirmed departures of Bansal and other key researchers, Meta's aggressive strategy appears to be yielding results.