Comment by Meta

This Advanced AI Scaling Framework outlines how Meta manages and prepares for Frontier AI capabilities that could lead to severe, large-scale outcomes. [...] For each Frontier AI, we outline anticipated capabilities, planned deployment (i.e., internal deployment, limited deployment, controlled deployment, closed release, or open release) [...] Our AI risk assessment process systematically evaluates potential risks associated with ongoing development and deployment of Frontier AI, documenting mitigation strategies and residual risks across a set of applicable risk domains. [...] We assess residual risk before deployment, taking into consideration the details of the risk assessment, the results of evaluations conducted throughout training, and the mitigations that have been implemented.