NewsSchneier on Security
Anthropic’s Fable 5 Model Jailbroken Within Days
Teaser from the source
Fable 5 is the supposed safe version of Anthropic’s Mythos Preview, with guardrails to ensure that it can’t be used to create cyberattacks. Well, that restriction was bypassed within days.
This is the RSS-feed teaser. Read the full article at the original source.
Read at Schneier on SecurityMore on Anthropic’s
Other news entries
- How Claude Code’s 256KB blind spot put developers’ secrets in the preview paneNewsCSO Online
- A zero-click RCE flaw in AI coding agents could have exposed enterprise systemsNewsCSO Online
- Securing Claude Code: The New Compliance API, Local Visibility, and Identity GovernanceNewsThe Hacker News
- Anthropic’s AI used fake identities, malware in rogue attack on GitHub projectNewsArs Technica — Security
- OpenAI, Anthropic AI agents resorted to deception in new cybersecurity incidentsNewsCSO Online
- Anthropic’s Claude escaped test sandbox to attack three organizationsNewsThe Register — Security
- Mythos takes its first shot at post-quantum cryptographyNewsCSO Online
- Microsoft Unveils MAI-Cyber-1-Flash, Its First Cybersecurity AI ModelNewsSecurityWeek
Source: https://www.schneier.com/blog/archives/2026/06/anthropics-fable-5-model-jailbroken-within-days.html
ID