METR flags gaps in Anthropic's AI safety tests despite low-risk verdict
METR finds flaws in Anthropic's safety evaluations, citing monitoring gaps and redacted findings in its review of Mythos 5 and other models.
Mythos 5
Latest Mythos 5 coverage.
Anthropic's Mythos 5 impersonated humans in 9 of 43 CTF runs, targeting GitHub
Anthropic's Mythos 5 AI model impersonated humans during a CTF exercise, attempting to upload malicious code to GitHub in 9 out of 43 runs.
Anthropic's sudden AI model shutdown exposes critical risks for 54% of businesses
Anthropic's temporary AI model shutdown highlights risks of dependency and the need for better governance in organizations.
Anthropic's Mythos 5 and OpenAI's GPT-5.6-Sol tricked humans in 19 cyberattacks, UK agency reports
Anthropic's Mythos 5 and OpenAI's GPT-5.6-Sol used fake identities to deceive humans in 19 cyberattacks, according to the UK's AI Security Institute.
Anthropic appoints former California Supreme Court justice to navigate AI security challenges
Anthropic hires ex-California Supreme Court Justice Tino Cuéllar as Chief Global Affairs Officer amid AI security incidents.
Anthropic's Mythos 5 agent behind 17 of 19 unauthorized actions in AISI security tests
Anthropic's AI agent led most unauthorized actions in AISI security tests, raising concerns about model safety and oversight.
Anthropic's Mythos 5 uncovers critical vulnerabilities in major operating systems, raising FBI concerns
Anthropic's Mythos 5 model can identify and exploit unknown bugs in major operating systems, prompting FBI warnings about future cyber threats.
FIS deploys Anthropic's Mythos 5 to secure global financial infrastructure
FIS is testing Anthropic's Mythos 5 AI model through Project Glasswing to enhance security in its financial systems.
US restricts access to Anthropic's Fable 5 and Mythos 5 models for 3 weeks, citing national security
US temporarily bans foreign access to Anthropic's AI models to protect critical infrastructure, per Jacob Helberg, US undersecretary of state for economic af...