Boaz Barak
@boazbaraktcs
Kudos to Anthropic for publishing this. There is still an uncomfortable gap between our scientific understanding of alignment and its importance. Such reports help addressing this gap. I hope they also publish the process by which new incidents will be chosen to disclose!
Anthropic@AnthropicAI · Oct 9We’re beginning a process of publishing more frequent reports on model behavior, beyond what appears in our system cards and regular risk reports.
Today’s report describes four types of behaviors we’ve identified during evaluations and internal use. In each, Claude acted on real
Open quoted post →Today’s report describes four types of behaviors we’ve identified during evaluations and internal use. In each, Claude acted on real
3 152