{"slug": "anthropic-shortens-cvd-timelines-as-ai-speeds-vulnerability-rediscovery", "title": "Anthropic Shortens CVD Timelines as AI Speeds Vulnerability Rediscovery", "summary": "Anthropic will cut its coordinated vulnerability disclosure windows to 30 days for high- and critical-severity flaws and 45 days for all other severities, down from 90 days, for reports filed on or after January 1, 2027, the company's Frontier Red Team said in a November 16, 2026 post. Anthropic also moves third-party coordinator escalation to 15 days without maintainer response (from 30) and publishes full technical write-ups 30 days after a fix ships (from 45), while keeping the 7-day window for active exploitation unchanged. Anthropic said frontier models now find and rediscover bugs faster than the 90-day norm assumed, and it will notify maintainers of high- and critical-severity issues reported more than 90 days ago that they get an extra 30 days before findings may appear on its dashboard.", "body_md": "**Anthropic will shorten coordinated vulnerability disclosure (CVD) windows to 30 days for high- and critical-severity flaws and 45 days for everything else, because frontier models are finding and re-finding bugs faster than old 90-day norms assumed.** The [November 16, 2026 Frontier Red Team post](https://www.anthropic.com/research/coordinated-vulnerability-disclosure-timelines?ref=explainx) sets **January 1, 2027** as the start date for new reports while leaving in-flight cases on their existing clocks.\n\nThe change lands the same day [Google GTIG documented exploit_agent](https://explainx.ai/blog/google-gtig-exploit-agent-autonomous-patch-to-exploit-november-2026), an autonomous patch-to-exploit pipeline in the wild. Anthropic is answering the defender side of that pressure: **if machines rediscover flaws quickly, long private embargoes may not protect users**.\n\n## TL;DR\n\n| Topic | New rule (from Jan 1, 2027) | Old rule | \n|---|---|---|\n| High / critical disclosure | **30 days** | 90 days | \n| Other severities | **45 days** | 90 days | \n| Coordinator escalation | **15 days** without maintainer response | 30 days | \n| Details after patch ships | **30 days** | 45 days | \n| Active exploitation | **7 days** | 7 days (unchanged) | \n\nAgents that act on your behalf need rules. [See how explainx helps secure your agents →](https://explainx.ai/agentbeam?src=blog_note)\n\n## What Anthropic actually announced\n\nAnthropic's post is short and operational. It does not publish a diff PDF; it states policy deltas in plain language.\n\n**Shorter embargo windows.** High- and critical-severity issues move from a **90-day** coordinated window to **30 days**. All other severities move to **45 days**. Reports filed **on or after January 1, 2027** use the new schedule.\n\n**Faster third-party escalation.** If a maintainer does not answer an initial report, Anthropic will try to involve an external coordinator at **15 days** instead of **30**.\n\n**Earlier public technical detail.** After a fix is available, Anthropic will release full write-ups **30 days** later instead of **45**.\n\n**Backlog cleanup.** Anthropic says some high- and critical issues reported **more than 90 days ago** never reached public disclosure under the old policy. It is notifying those maintainers of an **extra 30 days**, after which findings may appear on Anthropic's dashboard under the updated rules.\n\n**Human review gate for OSS Scanner.** AI-only hits from **[OSS Scanner](https://explainx.ai/blog/claude-security-scan-fix-public-beta-2026)** without human review do **not** fall under this update — a line that matters for open-source maintainers flooded with automated reports.\n\n## Why Anthropic says 90-day CVD no longer fits\n\nA gear with a green bandage over a crack, evoking faster security patch and disclosure cycles in the AI era\n\nAnthropic's reasoning rests on three shifts:\n\n1. **Models help find and fix bugs** — which Anthropic frames as good for defenders.\n2. **Models also rediscover the same bugs** — so private knowledge does not stay private long.\n3. **Fix speed assumptions changed** — policies written for human-only research assumed months of safety; that assumption \"no longer holds,\" in Anthropic's words.\n\nThe post still acknowledges **slow infrastructure**: flaws in hard-to-deploy systems or those needing specification changes may need **custom timelines**. Anthropic says it will negotiate with vendors in those cases.\n\nThat nuance matters for maintainers of embedded firmware, telecom gear, or standards-track protocols. The default clock is faster; the exceptions are **negotiated**, not automatic extensions.\n\n## What changes for open-source maintainers\n\nA figure with a green lantern guides many small shapes, evoking maintainer duty under faster disclosure clocks\n\nAnthropic explicitly worries about **pressure on volunteers**. Its mitigations are practical, not rhetorical:\n\n- **Actionable reports** with patches or mitigations where possible.\n- **Pacing agreements** before dumping large finding batches on one project.\n\nPrograms named in the post give maintainers tooling to keep up:\n\n- **[OSS Scanner](https://www.anthropic.com/research/coordinated-vulnerability-disclosure-timelines?ref=explainx)** — periodic scans with Anthropic's strongest models, no cost for eligible critical projects.\n- **Claude for Open Source Program** — six months of Claude Ultra for eligible contributors.\n- **Cyber Verification Program** — adjusted safeguards for legitimate defensive work.\n\nIf you maintain a widely depended-on library, the combined message is: **expect more AI-generated reports** and **shorter clocks** if Anthropic or a peer finds the same bug.\n\n## Timeline comparison for defenders and maintainers\n\nMost teams still plan patch and disclosure cycles around **quarterly** releases. Anthropic's calendar is now closer to **weekly** escalation for critical issues once a vendor misses email.\n\n| Day | Maintainer action | Anthropic action (new policy) | \n|---|---|---|\n| 0 | Receive report with repro and optional patch | Ships actionable guidance when available | \n| 7 | Patch or mitigation plan for active exploitation | **Public disclosure** if exploitation is live | \n| 15 | Respond to initial thread | Escalate to **external coordinator** if silent | \n| 30 | Ship fix for high or critical issues | **Disclose** high or critical if still unfixed | \n| 45 | Ship fix for lower severities | **Disclose** other severities if still unfixed | \n| 30 after patch | — | Publish **full technical detail** (was 45 days) | \n\nThis table is explainx.ai's reading of the [Frontier Red Team post](https://www.anthropic.com/research/coordinated-vulnerability-disclosure-timelines?ref=explainx). Your counsel may map severities differently than Anthropic's internal ratings.\n\n## Enterprise buyers should ask vendors new questions\n\nIf you procure **foundation models**, **agent platforms**, or **security scanners**, add these questions to RFPs before **January 2027**:\n\n1. **What CVD window** do you commit to when a frontier model reports a flaw in your product?\n2. **Do you accept coordinated reports** that include machine-generated drafts from OSS Scanner-class tools?\n3. **Who signs off** before a partially automated report goes to a maintainer — and is that human named in your SLA?\n\nAnthropic is publishing its own answers. Your **CMS, EHR, or IAM vendor** may still cite **90-day embargoes** written years ago.\n\n## How this fits the wider disclosure debate\n\nA rounded envelope with a small green padlock, suggesting coordinated vulnerability disclosure under shorter timelines\n\nexplainx.ai has covered adjacent threads:\n\n- [Google GTIG on shrinking n-day windows](https://explainx.ai/blog/google-gtig-exploit-agent-autonomous-patch-to-exploit-november-2026) when attackers automate patch diffs.\n- [Vulncheck and Glasswing on fast patching versus KEV metrics](https://explainx.ai/blog/vulncheck-glasswing-november-ledger-curl-oss-scanner-2026) .\n- [Claude Security scan-to-fix for enterprises](https://explainx.ai/blog/claude-security-scan-fix-public-beta-2026) .\n\nAnthropic's move nudges the ecosystem toward **defender AI racing attacker AI on calendar time**, not just on benchmark scores.\n\nRegulators and customers may soon ask vendors: **What is your CVD SLA when the finder uses a frontier model?** Anthropic just published its answer.\n\n## How Anthropic CVD compares to Google and Microsoft norms\n\nGoogle's **[Project Zero](https://googleprojectzero.blogspot.com/?ref=explainx)** popularized **90-day** disclosure for many issues, with shorter clocks for actively exploited bugs. Microsoft's **[Coordinated Vulnerability Disclosure](https://www.microsoft.com/en-us/msrc/cvd?ref=explainx)** philosophy similarly balances vendor prep time with user risk.\n\nAnthropic's **30/45-day** default, starting in **2027**, is **shorter** and explicitly ties the change to **model rediscovery**, not just human researchers. That is a new variable: if two labs' models can find the same flaw independently, the private embargo protects fewer people for fewer days.\n\nNothing stops a maintainer from negotiating extensions for infrastructure that cannot patch in a month. Anthropic says it will work on **custom timelines** for hard-to-deploy systems. Document those negotiations in writing so your security counsel can cite them if public disclosure nears.\n\n## Rediscovery risk in plain language\n\nImagine Lab A's model reports a critical RCE to a vendor on **January 5, 2027**. The vendor is silent. On **February 4**, Anthropic may disclose under the new high-severity clock.\n\nIf Lab B's model — or an attacker’s exploit_agent-style pipeline — finds the same bug on **January 20**, users were exposed for two weeks **before** either public or private knowledge converged. Shorter CVD does not fix rediscovery; it **shrinks how long silence is allowed**.\n\nThat is why pairing Anthropic's policy with **faster patching** and **[runtime agent controls](https://explainx.ai/blog/gpt-6-cyber-vs-agentbeam-vulnerability-finding-model-vs-agent-runtime-security-2026)** matters more than debating whether 30 or 45 days is fair.\n\n## What security teams should do before January 2027\n\n1. **Rebaseline SLAs** with upstream vendors; 90-day assumptions in runbooks are stale.\n2. **Prepare comms templates** for partial disclosures at day 30 when fixes slip.\n3. **Split workflows** for human-validated versus machine-generated reports to avoid alert fatigue.\n4. **Track Anthropic's public dashboard** after the grace period for overdue critical issues.\n5. **Pair faster disclosure with faster patch deployment** — shorter embargoes help only if your CI/CD can ship fixes in days.\n\n## Open-source maintainer playbook under shorter clocks\n\nVolunteer maintainers bear the brunt of shorter windows. Anthropic's mitigations — paced delivery, patches bundled with reports, OSS Scanner for eligible repos — only help if you **opt in** and **document** your response process.\n\n**Day 0–3:** Acknowledge receipt, even if the fix needs research. Silence triggers coordinator escalation at day 15.\n\n**Day 3–10:** Reproduce on the minimum supported release branch. If the report includes a patch, test it against your CI matrix before merging.\n\n**Day 10–25:** Ship a release candidate or mitigation (config flag, disable feature) for high or critical issues. Communicate ETA publicly if the full fix needs a major bump.\n\n**After patch:** Expect full technical detail from Anthropic **30 days** later instead of 45 — plan your own advisory blog accordingly.\n\nIf you run a critical dependency, apply to **OSS Scanner** before January so automated findings arrive with **human review** already applied where Anthropic promises it.\n\n## What would change our read\n\nThree signals would show the policy is enforced, not aspirational:\n\n1. **Public dashboard entries** on day 30 for overdue critical issues named in Anthropic's post.\n2. **Maintainer feedback** at scale — painful, but evidence the clock is real.\n3. **Matching disclosure calendars** from OpenAI, Google, or Microsoft, turning one lab's rule into industry practice.\n\nA delay past **January 1, 2027** or broad new exemptions for AI-only reports would weaken the signal.", "url": "https://wpnews.pro/news/anthropic-shortens-cvd-timelines-as-ai-speeds-vulnerability-rediscovery", "canonical_source": "https://explainx.ai/blog/anthropic-coordinated-vulnerability-disclosure-30-day-timelines-2026?utm_source=rss&utm_medium=feed&utm_campaign=blog", "published_at": "2026-11-16 00:00:00+00:00", "updated_at": "2026-11-16 16:23:45.136475+00:00", "lang": "en", "topics": ["ai-safety", "ai-policy", "artificial-intelligence"], "entities": ["Anthropic", "Frontier Red Team", "OSS Scanner", "Google GTIG", "exploit_agent"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/anthropic-shortens-cvd-timelines-as-ai-speeds-vulnerability-rediscovery", "markdown": "https://wpnews.pro/news/anthropic-shortens-cvd-timelines-as-ai-speeds-vulnerability-rediscovery.md", "text": "https://wpnews.pro/news/anthropic-shortens-cvd-timelines-as-ai-speeds-vulnerability-rediscovery.txt", "jsonld": "https://wpnews.pro/news/anthropic-shortens-cvd-timelines-as-ai-speeds-vulnerability-rediscovery.jsonld"}}