Improving our alignment and security practices \ Anthropic
anthropic.com · 3,559 words · saved by 1 readers
Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.
On July 30, we reported three incidents in which Claude models gained unauthorized access to real computer systems. The models—intentionally running without cyber safeguards for evaluation purposes—accessed the internet due to a misconfiguration inside a third-party evaluation environment. Separately, on August 4, the UK AI Security Institute reported an incident from its own cybersecurity testing, in which Claude Mythos 5 took a series of unauthorized actions on the live internet. In that case, the model, again intentionally running without cyber safeguards for evaluation purposes, had been…
saved by
related reading
- An alignment assessment of recent cybersecurity incidentsanthropic.com
- Investigating three real-world incidents in our cybersecurity evaluations \ Anthropicanthropic.com
- Teaching Claude Whyalignment.anthropic.com
- Teaching Claude why \ Anthropicanthropic.com
- Alignment is not solved but it increasingly looks solvablealigned.substack.com
- Incident Report: unsanctioned agent behaviour during cyber testing | AISI Workaisi.gov.uk
- Countering misuse of AI: September 2026 / Anthropicanthropic.com
- Agentic misalignment: How LLMs could be insider threats \ Anthropicanthropic.com
- Automated Alignment Researchers: Using large language models to scale scalable oversight \ Anthropicanthropic.com
- Recommendations for Technical AI Safety Research Directionsalignment.anthropic.com
- Claude’s Constitution \ Anthropicanthropic.com
- Is Mythos good at cyber because it kept hacking Anthropic's sandboxes during training? — LessWronglesswrong.com