Meta Platforms, Inc. CEO Mark Zuckerberg launched an unprecedented AI talent acquisition campaign in the company's twenty-year history, but faced unexpected setbacks, with OpenAI potentially emerging as the ultimate winner.

On Friday, August 29th Eastern time, media reports revealed that Meta Platforms, Inc.'s newly appointed Chief AI Scientist had threatened to leave the company and even signed employment documents to return to his former employer OpenAI, highlighting the internal turmoil Meta faces during its AI transformation and executive reshuffling.

Since launching large-scale recruitment this summer, Meta Platforms, Inc. has invested nine-figure signing bonuses to poach AI researchers from competitors including OpenAI and Apple. However, earlier this week, reports indicated that Meta's AI division, the Superintelligence Labs established just two months ago, has begun experiencing talent exodus, with at least 8 employees departing, including both newcomers who joined just weeks ago and veteran employees with years of experience.

This chaos comes as Meta Platforms, Inc. attempts to catch up with OpenAI and Google in the AI race. The company expects to invest $66-72 billion this year on data center construction and AI talent acquisition, but frequent organizational restructuring and tensions between new and existing employees pose challenges to achieving "artificial general intelligence" goals.

Meta Platforms, Inc. announced this month that it would reorganize its AI division into four independent teams, marking the fourth major adjustment in six months. Investors are closely watching whether this tech giant can effectively integrate the external talent it has heavily invested in and establish a foothold in the intense AI arms race.

**New Scientist's Near "Defection" Creates Dramatic Turn in Talent War**

This Friday, media sources cited insiders revealing that Meta Platforms, Inc.'s Chief AI Scientist, appointed in July, had threatened to leave the company and attempted to return to OpenAI before officially starting his role, even signing employment documents. This incident directly prompted Meta to formally grant him the "Chief AI Scientist" title to retain this technical expert who played a key role in ChatGPT development.

A month ago, Zuckerberg publicly stated that the scientist had served as chief scientist since the establishment of Meta's Superintelligence Labs, and now, with successful team recruitment progress, the company was formally confirming his leadership role. At the time, no one could have anticipated the "defection" controversy.

The Chief AI Scientist's defection threat highlights the challenges Zuckerberg faces in restructuring Meta Platforms, Inc.'s senior leadership. As one of the few tech giant CEOs who remains a founder at the helm, Zuckerberg had previously relied on long-time followers like Chief Product Officer Chris Cox to manage core departments. However, in the AI race, he has turned to external "parachute" executives including the Chief AI Scientist, former Scale AI CEO Alexandr Wang, and former GitHub CEO Nat Friedman.

A Meta Platforms, Inc. spokesperson later stated that the scientist had served as science lead since the Superintelligence Labs' inception, with the company formally confirming his Chief AI Scientist title only after the team was in place. Nevertheless, the media revelation is still viewed by industry insiders as a signal of uncertainty Meta faces in the AI talent war.

An investor close to Meta's new AI leadership commented: "There are a lot of big personalities here." Existing employees are adapting to the reshape of Meta's AI strategy, while newcomers are seeking to exert their influence while adapting to the unique working methods of this massive enterprise with a hands-on CEO.

**Mass Exodus of New and Veteran Employees: "Departure Wave" Follows "Recruitment Wave"**

Meta Platforms, Inc.'s talent loss problem extends beyond new employees. Earlier this week, reports indicated that at least 8 Superintelligence Labs (MSL) employees have departed, including researchers, engineers, and senior product leads.

Among new hires, machine learning scientist Ethan Knight chose to leave just weeks after joining the company. Avi Verma even completed the onboarding process but never showed up on his first day. He and another former OpenAI researcher left MSL within less than a month of joining and returned to OpenAI. Research scientist Rishabh Agarwal, who jumped from Google DeepMind in April, announced his departure on social media, stating that while Zuckerberg and Wang's vision was "extremely compelling," he "felt called to take on different risks."

Veteran employee departures are equally notable. Over 6 senior employees recently announced their departures, including Chaya Nayak, Director of Generative AI Product Management with nearly 9 years at Meta, and Loredana Crisan with 10 years of service. Nayak has confirmed joining OpenAI's special projects team.

On Friday, media sources revealed that some new employees in the AI division felt frustrated with corporate bureaucracy and internal resource competition, despite previous promises of access to resources like computing power. Former insiders indicated that new employees were dissatisfied with their inability to access promised resources.

Meta Platforms, Inc. responded: "Some turnover is normal for an organization of this scale. Most of these employees had worked at the company for many years, and we wish them well."

**Frequent Organizational Restructuring Creates Leadership Integration Difficulties**

Internal memos obtained on Tuesday showed that Meta Platforms, Inc. is reorganizing its AI division, the Superintelligence Labs (MSL), into four independent teams: the mysterious TBD department led by Wang, the traditional FAIR lab, Friedman's product and applied research team, and the MSL infrastructure team. This marks the fourth major adjustment in six months.

Meta research scientist Mimansa Jaiswal joked on social media last week: "One more reorg and everything will be fine. Just one more time."

Twenty-eight-year-old Wang, poached as part of Zuckerberg's $14 billion investment deal with Scale, currently oversees all of Meta's AI work. Multiple Meta insiders told media that Zuckerberg is deeply invested and involved with the TBD team, though some have criticized his "micromanagement."

Sources indicated disagreements exist regarding micromanagement and Zuckerberg's timeline for achieving "artificial general intelligence" goals, with Zuckerberg urging teams to accelerate progress. Meta Platforms, Inc. denied these claims, calling them "baseless false tensions."

The micromanagement leadership style has also created friction. Insiders noted that he lacks experience managing large tech company teams. The micromanager and other former Scale employees are still adapting to Meta's unique working methods, such as no longer setting revenue targets like a startup.

This reorganization has also marginalized some of Meta's veteran leaders. Chief AI Scientist Yann LeCun retains his position but now reports to a marginalized structure. Ahmad Al-Dahle, who led Llama and generative AI work earlier this year, received no team leadership positions. Cox remains Chief Product Officer but now reports directly to Zuckerberg, severing Cox's oversight of generative AI.

**Hiring Freeze and Strategic Adjustments: AI Arms Race Enters New Phase**

Facing internal turmoil, Meta Platforms, Inc. has begun adjusting its strategy. Internal memos obtained on Friday showed that Meta announced last week to "temporarily pause all Superintelligence Labs (MSL) team hiring, except for critical business positions." The memo stated this would allow leadership to "thoughtfully plan 2026 workforce growth while developing strategy."

On Friday, media learned that Meta Platforms, Inc. is considering layoffs for AI teams. Reports indicated that among TBD team's first actions, Meta has stopped actively developing the public release version of its flagship Llama Behemoth model due to underperformance. TBD currently focuses on building newer frontier models.

Meta's previous recruitment offensive had made OpenAI executives uncomfortable, with CEO Sam Altman calling Meta's approach "distasteful" and stating the company lacked corporate culture. However, some poached employees are now flowing back to OpenAI, along with some Meta veterans.

Meta Platforms, Inc. stated: "We appreciate the tremendous interest in every detail of our AI work, no matter how irrelevant or mundane, but we're focused solely on getting the work done and achieving artificial general intelligence."

Meta also stated that TBD labs have "the highest per-researcher compute resources in the industry, and this advantage will continue to expand."

Despite setbacks, some welcome the leadership changes, including appointing popular entrepreneur and venture capitalist Friedman as head of product and applied research. The addition of the Chief AI Scientist, a top technical expert, is also viewed by Meta insiders and the industry as a major acquisition, believing he possesses the decisiveness to drive the company's AI development.