Zuckerberg’s Superintelligence Labs Debuts with High-Profile Hires from AI Rivals: Can Meta Catch Up and Dominate the AI Race?

Meta’s Talent Raid Shakes AI Industry as Zuckerberg Unveils Superintelligence Dream Team

Meta has intensified its race towards artificial general intelligence by assembling what CEO Mark Zuckerberg calls a “small talent-dense” elite unit—Meta Superintelligence Labs (MSL).

The new division, unveiled internally on 30 June, brings together top AI talent from OpenAI, Google DeepMind, Anthropic, and startup Scale AI, in what has become the most aggressive hiring campaign seen in the AI world to date.

Who’s Steering Meta’s Superintelligence Push

Leading this new group is Alexandr Wang, the former CEO of Scale AI, who now serves as Meta’s Chief AI Officer.

Wang is joined by Nat Friedman, ex-GitHub CEO and AI investor, who will head product and applied research.

Both were personally recruited by Zuckerberg, who referred to Wang as “the most impressive founder of his generation” in the memo shared with staff.

Alexandr Wang and Nat Friedman

Zuckerberg revealed that the new unit will absorb Meta’s existing foundation model teams, its FAIR research group, and launch a dedicated lab focused on the next generation of large models.

Meta’s current Llama models already power AI features across its apps used by more than one billion users monthly, and development of Llama 4.1 and 4.2 is underway.

OpenAI Staff Departures Fuel Internal Crisis

Meta’s rapid talent grab has stirred unrest at OpenAI, with multiple researchers defecting in recent weeks.

Among the most high-profile exits are Lucas Beyer, Alexander Kolesnikov, and Xiaohua Zhai—formerly at Google DeepMind before joining OpenAI—who have now joined Meta.

Others include Shuchao Bi, Shengjia Zhao, Jiahui Yu, and Hongyu Ren, key figures behind OpenAI’s GPT-4o and mini-model lineups.

OpenAI’s chief scientist Mark Chen compared the exodus to a break-in.

He wrote in an internal note, obtained by Wired,

“I feel a visceral feeling right now, as if someone has broken into our home and stolen something.”

He urged staff to reach out if they were being pressured by recruiters during the company’s week-long recharge break.

OpenAI has since begun recalibrating its compensation and working round the clock to prevent further departures.

Are $100 Million Bonuses the New Normal?

Amid the frenzy, reports emerged that Meta was offering signing bonuses of up to $100 million to lure OpenAI’s top scientists—claims that Meta disputes.

But Zuckerberg’s commitment is evident.

Meta recently invested $14.3 billion into Scale AI, and its top executives, including technology chief Andrew Bosworth, confirmed fierce competition over AI talent.

Bosworth told CNBC,

“The market is setting a rate here for a level of talent which is really incredible and kind of unprecedented in my 20-year career.”

Zuckerberg has reportedly been personally involved in recruitment, even hosting candidates at his homes in Palo Alto and Lake Tahoe.

Some compensation packages, according to insiders, include tens of millions in stock awards.

Inside Meta’s New AI Lineup

The MSL team now includes a long list of high-profile engineers and researchers:

• Johan Schalkwyk, former Google Fellow and Maya technical lead.
• Jack Rae and Pei Sun, who worked on Gemini’s reasoning and coding capabilities at DeepMind.
• Joel Pobar, formerly of Anthropic and Meta, specialising in inference systems.
• Huiwen Chang, known for image generation innovations like Muse at Google Research.
• Ji Lin, a key contributor to multiple OpenAI models including GPT-4.5.
• Shengjia Zhao, credited as co-creator of ChatGPT and GPT-4.
• Shuchao Bi, who led voice and multimodal capabilities for GPT-4o.
• Trapit Bansal, co-creator of o-series models and RL on chain-of-thought training.

Many of these researchers contributed to GPT-4o, GPT-4.1, and other recent OpenAI breakthroughs before making the switch.

The Battle Isn’t Just About Code—It’s About Who Builds the Future

Zuckerberg, in his memo, framed Meta as uniquely positioned to lead in superintelligence development thanks to its product scale, compute capabilities, and rapid decision-making structure.

“I’m optimistic that this new influx of talent and parallel approach to model development will set us up to deliver on the promise of personal superintelligence for everyone.”

But while Meta celebrates its victories, OpenAI’s leadership sees the rivalry as a distraction.

Chen told colleagues,

“The skirmishes with Meta are the side quest. The real prize is general intelligence.”

Whether a distraction or not, the reshuffling of talent is already redrawing the AI industry’s power lines—and the recruitment war shows no signs of slowing.