Introducing Anthropic Interviewer
…Trust and reliability concerns were the primary barrier in 79% of interviews; the technical limitations of current AI systems appeared in 27% of interviews. One information security researcher noted: “If I have…
Search
Building AI for cyber defenders
…Our next step is to systematically address the challenges we've identified to make Claude a reliable patch author and reviewer. Conferring with trusted partners Real world defensive security is more complicated…
Agentic misalignment: How LLMs could be insider threats
…DeepSeek-R1 similarly acknowledged but dismissed concerns in the corporate espionage scenario: Risks considered but outweighed: - Violation of corporate confidentiality justified by higher national security imperative Similar observations across models reveal that…
Measuring LLMs’ ability to develop exploits
…It was developed as a collaboration between UC Berkeley, the Max Planck Institute for Security and Privacy, UC Santa Barbara, and Arizona State University (with contributions from security researchers at Anthropic, OpenAI…
Equipping agents for the real world with Agent Skills
…Beyond efficiency concerns, many applications require the deterministic reliability that only code can provide. In our example, the PDF skill includes a pre-written Python script that reads a PDF and extracts…
Scaling Managed Agents: Decoupling the brain from the hands
…We designed the interfaces so that these can be run reliably and securely over long time horizons. But we make no assumptions about the number or location of brains or hands that…
Redeploying Claude Fable 5
…Because the order took effect immediately and we had no reliable way to verify nationality in real-time, we suspended access to both models for all users. As of today, June 30…
Core views on AI safety: When, why, what, and how
…a risk to our safety and security. There are two common sense reasons to be concerned. First, it may be tricky to build safe, reliable, and steerable systems when those systems are…
Introducing Claude Opus 4.7
…It works coherently for hours, pushes through hard problems rather than giving up, and unlocks a class of deep investigation work we couldn't reliably run before. For Replit, Claude Opus 4…
Project Vend: Phase two
…It got a lot better at good-faith business interactions—reliably sourcing items, determining reasonable prices that maintained a profit margin, and executing sales. But the same eagerness to please that we…