Mythos-Class AI Models
Anthropic confirms plans to release Mythos-class models to the public in the coming weeks, after addressing initial security risks.
HalluSquatting is an attack where researchers pre-compute fake repository, package, or skill names that AI coding agents predictably invent, register those names first, and load them with malicious instructions.
Anthropic confirms plans to release Mythos-class models to the public in the coming weeks, after addressing initial security risks.
A Google security engineer has been accused of using confidential search trends to make over $1.2 million on the prediction marketplace Polymarket.
CrowdStrike has dismantled the Glassworm botnet, which infected hundreds of open-source software with malware, in a coordinated effort with Google and Shadowserver.
Edamame's new platform aims to catch AI coding agents going off the rails by detecting code drift and attack patterns in real-time.
Security researchers discovered a bug chain in Zapier that could have granted access to millions of user accounts and connected systems, but the issues have been fixed.
Russia is conducting daily hybrid attacks against the UK, targeting critical infrastructure, democratic processes, and public trust, according to GCHQ director Anne Keast-Butler.
SIEM helps MSPs reduce noise and stop threats faster by providing a centralized view of activity across the entire environment and automatically correlating related events.
Geordie, an AI security and governance startup, has raised $30 million in a Series A funding round to enhance its platform and expand operations.
A House subcommittee will hold a hearing on June 4 to discuss the impact of frontier artificial intelligence models on cybersecurity threats.