Reports suggest that Meta is contemplating a shift from open to closed AI models, potentially altering its technological approach.
Potential Changes in Meta's Strategy
A recent meeting among senior members of Meta, including newly appointed AI chief Alexandr Wang, confirmed discussions about pausing the release of their open-source Behemoth in favor of a proprietary version. This move would mark a significant change in Meta's policy, which has traditionally embraced open development.
Recent Changes in Meta's Structure
Since expanding its superintelligence division, Meta has adopted more aggressive hiring strategies, offering substantial compensation to attract top talents from firms like OpenAI and Google. Wang's appointment as AI chief was part of this initiative, and the team's first meeting in Menlo Park focused on keeping their work private, allowing them to concentrate on superintelligence development.
Looking Ahead to the Initiative
In light of these shifts, Zuckerberg announced plans to invest hundreds of billions in AI compute infrastructure, aiming to operationalize a supercluster by next year. He emphasized that the team's work would be pivotal in advancing towards superintelligence, signifying a new phase in the company's AI development.
Meta's shift to closed AI models raises questions about the future of open-source within the company, yet highlights its determination to compete with industry leaders.
