Anthropic has announced the transfer of the Petri project to Meridian Labs to foster the tool's independent growth within the open-source community.
Developments
Petri is a tool developed by Anthropic to evaluate the alignment and safety of AI models. Alongside the transfer, Anthropic and Meridian Labs have also rolled out a major update, improving the adaptability, realism, and depth of evaluations in Petri. This move transitions Petri away from single-company control, turning it into a shared asset for the community.
Why It Matters
Open-sourcing and transferring AI safety tools is a crucial trend for building trust. For AI development teams in Vietnam, Petri offers a standardized framework to test whether their models are operating as intended and safely. Meridian Labs' stewardship promises to bring more diverse contributions from the global developer community.