The FMF is currently advancing shared understandings of AI safety and security across four workstreams:
