Developing nuclear safeguards for AI through public-private partnership
… We have already deployed this classifier on Claude traffic as part of our broader system for identifying misuse of our models. …
Search
2 sources covering this — show 1 more
Building safeguards for Claude
… This helps us spot behaviors that might appear violative only in aggregate, such as automated influence operations and other large-scale misuses. …
Inviting hard questions
… But many also hold serious concerns. …
Mapping AI-enabled cyber threats: Insights from the LLM ATT&CK Navigator
… It also serves as a guide to how threat actors are likely to misuse increasingly capable models in the near future, giving defenders a chance to get ahead of them. What we learned from this and other analyses directly shapes how we build Claude to prevent such misuse. …
Our position on open-weights models
… Protectionist bans would not address my most serious national security concerns. …
LLMs and biorisk
… Across all models, our Safeguards team monitors activity on our platforms for signs of misuse, including dangerous applications of biology. This monitoring provides insight into patterns of misuse and helps us determine if and when to take enforcement action. …
Statement on the US government directive to suspend access to Fable 5 and Mythos 5
… Anthropic’s posture with respect to Fable’s safeguards, as laid out in our launch blog post , is the following: We have instituted strong safeguards that greatly reduce the likelihood that Fable is misused for tasks related to cybersecurity among others . …
Redeploying Claude Fable 5
… These prodigious cybersecurity capabilities make it uniquely attractive to malicious actors who wish to misuse it in cyberattacks. …
Improving Fable 5 Safeguards
… Nevertheless, we chose to make this tradeoff because the cost of Fable being misused in a dual-use domain like biology could potentially be catastrophic . …