technologyScore 30/100Watch
Amodei: Models approaching ASL3 (serious national security risk) threshold in 2025; will require mitigation measures for chemical/biological/radiological weapon design information
Dario Amodei· Anthropic· AI· 2026-07-18· about Anthropic
“ASL3 is characterized um and you know we focus very much on the national security side very kind of serious risk that are out of proportion to the risks that normal technologies have... And we believe we have a story for how to deploy those kinds of models safely by you know removing their ability to do this very narrow range of dangerous tasks without compromising their commercial viability... I think we're approaching that. We may actually hit that uh uh uh this year.”
Why it matters
Reaching ASL3 will trigger mandatory safety measures and government testing. This signals upcoming regulatory escalation and formalization of AI safety protocols. Companies that cannot meet ASL3 deployment requirements may face government restrictions.
Investment implication
AI safety companies, security consulting firms, and government contractors supporting AISI testing will see accelerated demand. Commercial AI deployment may face delays pending ASL3 clearance. Regulatory arbitrage opportunities for compliant vs. non-compliant vendors.