Meta's AI Overhaul: Chasing Superintelligence Amid Talent Wars and Ethical Risks

Meta Platforms Inc. has announced a significant reorganization of its artificial intelligence (AI) division, restructuring its newly formed unit into four specialized teams to accelerate the development of "superintelligence." The move, outlined in an internal memo by new Chief AI Officer Alexandr Wang, reflects the company’s intensified efforts to compete with rivals like OpenAI and Google. The restructured unit, now named Meta Superintelligence Labs (MSL), aims to consolidate and streamline research, product development, and infrastructure work in pursuit of AI systems that can outperform humans in complex tasks [1].

The four new groups under MSL are the TBD Lab, led by Wang, which oversees the company’s large language models including the Llama series; the Fundamental AI Research (FAIR) lab, which has existed for over a decade and focuses on long-term research; Products and Applied Research, led by former GitHub CEO Nat Friedman, which is tasked with integrating AI into consumer-facing applications; and MSL Infra, focused on building the costly infrastructure necessary for AI development [1]. The reorganization does not include layoffs, according to a source familiar with the matter [1].

This restructuring marks Meta’s fourth major reorganization of its AI teams in six months, a sign of the company’s urgency to catch up with industry leaders. CEO Mark Zuckerberg has been deeply involved in attracting top-tier talent, offering multi-million-dollar compensation packages to recruit former executives and researchers from companies such as Scale AI, GitHub, and Google DeepMind [1]. The company has also raised its 2025 capital expenditure forecast to $66 billion to $72 billion, with further spending anticipated in 2026 to support AI infrastructure and talent acquisition [2].

The MSL initiative follows a period of instability in Meta’s AI leadership and a poor reception of its latest open-source Llama 4 model. The company’s previous AI structure included three teams—FAIR, an AI products group, and the AGI foundations team—which has now been dissolved, with its former leaders shifting to strategic MSL roles [1]. Meanwhile, Meta has faced criticism over privacy concerns related to its AI assistant, with reports of potential unauthorized tracking of Android users and the public sharing of user prompts [4].

From a market perspective, marketers and investors are closely watching MSL’s progress. Many advertisers have expressed concerns over transparency and control in Meta’s AI-driven advertising tools, such as Advantage+, which they describe as an “AI black box” due to its opaque decision-making process [3]. There is also a demand for privacy-safe AI tools that can infer user demographics and intent without compromising data privacy. Additionally, brands are seeking AI-powered creative guidance that can predict successful campaign strategies rather than merely analyzing past performance [3].

Despite Meta’s aggressive hiring and investment strategy, skepticism remains. Critics have questioned the sustainability of the company’s high compensation packages and the potential ethical risks of pursuing superintelligence. Elon Musk has publicly criticized Meta’s approach, arguing that xAI can attract top talent without unsustainable salary offers [7]. Moreover, Meta’s pivot away from its open-source AI model—once a key differentiator from competitors like OpenAI and Google—has raised concerns about monopolization and reduced innovation access for smaller players [7].

Zuckerberg has emphasized that the ultimate goal of MSL is to deliver “personal superintelligence,” where AI systems are deeply integrated into individual lives to enhance productivity and creativity. This vision, however, faces practical and ethical challenges, including workforce displacement and the potential for AI to amplify existing biases. Analysts suggest that the success of MSL will depend on its ability to demonstrate real-world applications and competitive advantages over existing AI platforms [3].

Source:

[1] Meta Restructures AI Group Again in Pursuit of Superintelligence (https://www.bloomberg.com/news/articles/2025-08-19/meta-restructures-ai-group-again-in-pursuit-of-superintelligence)

[2] Meta plans fourth restructuring of AI efforts in six months (https://www.reuters.com/business/meta-plans-fourth-restructuring-ai-efforts-six-months-information-reports-2025-08-15/)

[3] Here's what marketers want from Meta's Superintelligence Labs (https://digiday.com/marketing/heres-what-marketers-want-from-metas-superintelligence-labs/)

[4] Meta Eyeing Fourth Revamp of AI Teams in Six Months (https://www.pymnts.com/meta/2025/meta-eyeing-fourth-revamp-of-ai-teams-in-six-months/)

[5] Meta is shaking up its AI org, again (https://techcrunch.com/2025/08/19/meta-is-shaking-up-its-ai-org-again/)

[6] Meta Restructures AI Group Again in Pursuit of (https://finance.yahoo.com/news/meta-restructures-ai-group-again-163342766.html)

[7] Meta Superintelligence Labs: What We Know So Far (https://builtin.com/artificial-intelligence/meta-superintelligence-labs)