# Meta Superintelligence Labs Meta's advanced AI research division focused on developing superintelligent AI systems, announced by [[Mark Zuckerberg]] as part of Meta's escalation in the AI arms race. The lab represents Meta's bet that the path to [[AGI]] and beyond runs through massive compute investment and open-weight model release — a strategy that puts Meta in direct tension with the safety-first approach of [[Anthropic]] and the closed-model strategy of [[OpenAI]]. Meta's AI approach is distinctive in the vault's framework because it combines frontier capability ambitions with an open-source philosophy that complicates export control and safety governance regimes. If the most capable models are released openly, the entire architecture of model-level restrictions — from the [[Anthropic Model Export Directive]] to [[CAISI]] evaluations — becomes harder to enforce. ## See Also - [[Mark Zuckerberg]] — Meta's CEO driving the AI strategy - [[AGI]] — the stated goal - [[AI Safety]] — the governance challenge open-weight models create - [[Silicon Valley AI Ecosystem]] — the competitive landscape