This is false. Our ASL-4 thresholds are clearly specified in the current RSP—see “CBRN-4” and “AI R&D-4″. We evaluated Claude Opus 4 for both of these thresholds prior to release and found that the model was not ASL-4. All of these evaluations are detailed in the Claude 4 system card.
The original commitment was (IIRC!) about defining the thresholds, not about mitigations. I didn’t notice ASL-4 when I briefly checked the RSP table of contents earlier today and I trusted the reporting on this from Obsolete. I apologized and retracted the take on LessWrong, but forgot I posted it here as well; want to apologize to everyone here, too, I was wrong.
This is false. Our ASL-4 thresholds are clearly specified in the current RSP—see “CBRN-4” and “AI R&D-4″. We evaluated Claude Opus 4 for both of these thresholds prior to release and found that the model was not ASL-4. All of these evaluations are detailed in the Claude 4 system card.
The thresholds are pretty meaningless without at least a high-level standard, no?
The RSP specifies that CBRN-4 and AI R&D-5 both require ASL-4 security. Where is ASL-4 itself defined?
The original commitment was (IIRC!) about defining the thresholds, not about mitigations. I didn’t notice ASL-4 when I briefly checked the RSP table of contents earlier today and I trusted the reporting on this from Obsolete. I apologized and retracted the take on LessWrong, but forgot I posted it here as well; want to apologize to everyone here, too, I was wrong.