Anthropic Cuts Disclosure Deadline to 30 Days For Serious Bugs
At a glance
- Severity
- Medium
- Used in attacks
- No flaws named
- Reported by
- 1 outlet
Anthropic has cut the time vendors get to fix serious flaws found by its Claude models before it makes them public from 90 days to 30. The change applies to vulnerabilities Anthropic reports to vendors and maintainers, not generally to bugs reported to Anthropic by external researchers.
The company's Frontier Red Team announced the change in a November 16 post, with the new timelines applying to vulnerabilities reported on or after January 1, 2027. Reports already in progress then will keep their existing deadlines.
Anthropic said disclosure norms, including its own, had assumed fixes moved at human speed and that independent rediscovery was unlikely, but that "neither assumption holds today." It said bugs one model could find were increasingly likely to be found by another.
What Changes in Anthropic's Disclosure Policy
Under the updated policy, high- and critical-severity vulnerabilities will have a 30-day disclosure deadline, while all other issues will have 45 days. Anthropic will aim to escalate to an external vulnerability coordinator after 15 days without a maintainer response, rather than 30.
The company will also wait 30 days after a patch becomes available before publishing full technical details, down from 45 days. The seven-day disclosure window for vulnerabilities under active exploitation remains unchanged.
Anthropic said it would work with vendors on a remediation timeline for flaws in hard-to-deploy software such as critical infrastructure, or flaws that need specification or RFC changes. The policy will not cover AI-discovered findings that have not received human review, such as those from its OSS Scanner.
Acknowledging that shorter timelines could put more pressure on maintainers, Anthropic said it would keep sending actionable guidance with each report, plus a patch or mitigation where possible.
Read more on vulnerability disclosure: AI Cuts Vulnerability Discovery from Nine Months to 90 Hours
How Anthropic's Deadlines Compare With Project Zero's
The change puts Anthropic's serious-bug deadline ahead of Project Zero's schedule. Google's team, which now publishes details 90 days after reporting a bug whether or not it is fixed, will cut that to 60 days on February 2, 2027 and 30 days on January 3, 2028.
The approaches also differ on severity. Anthropic will set different disclosure windows based on severity, while Project Zero argued that a vulnerability's severity does not affect how likely attackers are to find it.
On the same day, Google's Threat Intelligence Group reported finding an exposed server hosting an attacker framework, dubbed exploit_agent, built to research vulnerabilities and weaponize them.
Anthropic said it was now giving maintainers 30 more days on high- and critical-severity bugs that had passed its current 90-day deadline, before publishing them on its disclosure dashboard.
Originally published by Infosecurity Magazine. © Infosecurity Magazine.
Fastnexa security experts
Dealing with this in your own company?
If this story touches software, suppliers or systems you use, a Fastnexa security expert can tell you what it means for you and what to do first.
Think you’ve already been hit? Don’t wait on a form: call or WhatsApp +1 (732) 454 2616. We reply within 1 hour, 24/7. Emergency help →
Coverage
One outlet has carried this so far.
2026-11-17 15:30 UTC
Related stories
- Cloud keys leaked in Docker Hub images are stolen within seconds
BleepingComputer · 2026-11-17
- Researchers Claim Two RCE Exploits That Bypass Telegram’s New Android Defenses
The Hacker News · 2026-11-17
- Chinese state group developing automated attack tool for cyber espionage, Google says
The Record · 2026-11-17
- Cursor Traces Malicious Claude Code Extension to Third-Party Breach
Dark Reading · 2026-11-17
- AI Framework Autonomously Weaponized Flaws in Real Attacks, Google Says
SecurityWeek · 2026-11-17