Loading story…
Anthropic launched OSS Scanner, a free, opt-in service that uses its strongest AI models to repeatedly audit eligible open-source projects and email potential vulnerabilities directly to maintainers. Eligibility prioritizes established, widely used projects important to infrastructure and user security, particularly those exposed to remote attacks. Enrollment requires verified core-maintainer status, project configuration, and build instructions. The service builds projects in isolated virtual machines with network access, then disables Internet access before auditing. Reports include reproduction steps and proposed patches where available, but receive no human review before delivery.
In an assessment of 97 high- and critical-severity findings across 48 projects, penetration testers determined that 85 met Anthropic’s coordinated disclosure criteria, 11 were genuine but duplicative, and one was invalid. Maintainers remain responsible for validating findings and proposed fixes and prioritizing remediation; Anthropic warns that reports may overstate severity or misunderstand security assumptions. Projects can supply testing guidance, pause automated reports, or opt out. Unvalidated findings carry no mandatory 90-day disclosure deadline, although findings subsequently confirmed through Anthropic’s human-reviewed program may enter its disclosure process.
See affected versions and whether adversaries are exploiting it.
5 events from the most recent confirmed update back to the earliest known activity.
Anthropic unveiled the Critical Infrastructure Defense Program as part of its Cyber Mission to protect critical infrastructure and open-source software against AI-enabled cyber threats. The initiative aims to accelerate vulnerability fixes and explore secure architectures and coding practices.
Anthropic introduced OSS Scanner to repeatedly scan eligible open-source projects and email suspected vulnerabilities, reproduction instructions, and proposed fixes when available. Findings are delivered without human review, leaving maintainers responsible for validation; unvalidated reports carry no mandatory 90-day disclosure deadline.
During testing, OSS Scanner chained vulnerabilities into working unauthenticated remote-code-execution exploits against real software projects. Curl maintainer Daniel Stenberg reported multiple issues worth fixing, while wolfSSL’s Todd Ouska said 72 of 74 findings were valid and five became CVEs.
Penetration testers assessed 97 high- and critical-severity findings from an early version of OSS Scanner across 48 projects. Anthropic reported that 85 met its coordinated disclosure criteria, 11 were genuine but duplicative, and one was invalid.
Anthropic’s earlier security research included work in which Claude uncovered 22 vulnerabilities in Firefox.
Vulnerabilities, threat actors, malware, products, organizations, and breaches Mallory has linked to this story.
See whether adversaries are exploiting this yet, and where the affected versions run in your environment.
8 references tracked. Mallory keeps watching after this page renders.
securityaffairs.com
Open sourcethehackernews.com
Open sourcehelpnetsecurity.com
Open sourcecyberaccord.com
Open sourcecybersecuritynews.com
Open sourceanthropic.com
Open sourcegithub.com
Open sourcegoogle.github.io
Open sourceMap indicators from this story to your assets and identify affected systems in minutes.
Every observed campaign, victim, and pivot linked to actors named in this story.
Malware, exploits, and IOCs connected to the activity described here.
YARA, Sigma, and Snort rules deployed to your SIEM as soon as they’re published.
Get matching new stories delivered to your team as they break — not the next morning.
Ask questions about this story and take action on the answers.
Other stories Mallory is tracking that overlap on entities, sources, or topic labels.
Anthropic announced the **Anthropic Cyber Mission**, a program combining Claude models, its engineers and threat research with outside cybersecurity expertise to identify and fix vulnerabilities in critical infrastructure and open-source software. The company said it has offered frontier models and technical support to more than half of U.S. states and large critical infrastructure operators, supporting code scanning, patching, incident response and red teaming. The initiative also includes a **free, opt-in scanning service for open-source projects**, providing periodic vulnerability reports with proofs of concept, explanations and suggested patches. Anthropic cautioned that reports may contain inaccuracies, making validation important before acting on findings or applying fixes. Its longer-term goals include automating most vulnerability triage and patching and developing new security architectures and coding standards.
Oct 10, 2026
Anthropic unveiled **Claude Mythos Preview**, an unreleased AI model it says can autonomously discover and exploit severe software vulnerabilities across major operating systems, browsers, open-source projects, and some closed-source targets. The company said the model uncovered thousands of high-severity flaws, including long-lived bugs in **OpenBSD**, **FFmpeg**, Linux kernel privilege-escalation chains, and **`CVE-2026-4747`**, a FreeBSD NFS remote code execution flaw that could enable unauthenticated root access. Anthropic withheld broad release, citing offensive cyber risk, and instead launched **Project Glasswing**, a gated program for roughly 40 to 50 partners such as AWS, Apple, Cisco, Cloudflare, Google, JPMorgan Chase, Microsoft, Mozilla, NVIDIA, and Palo Alto Networks to validate findings, patch affected software, and study defensive uses. Independent and industry assessments broadly agreed Mythos marks a significant advance in AI-enabled cyber capability, though several researchers questioned how much of Anthropic’s headline claims can yet be verified through public CVEs and warned that similar results may be reproducible with cheaper or open models plus strong tooling. The UK AI Security Institute found Mythos achieved a **73%** success rate on expert capture-the-flag tasks and completed a full 32-step simulated enterprise attack in 3 of 10 runs, while Anthropic later reported coordinated disclosure activity spanning **1,596 vulnerabilities across 281 open-source projects** and partners identifying more than **10,000** high- or critical-severity candidates. Governments, financial regulators, and CISO groups in the US, UK, Europe, Canada, and Japan responded with briefings and warnings that AI is compressing the gap between vulnerability discovery and weaponization, leaving remediation, patch governance, and defensive automation as the main bottlenecks.
Aug 31, 2026
Anthropic’s Claude Mythos Preview drew attention for reported advances in vulnerability discovery and exploit development, but subsequent open-source maintainer feedback challenged the practical value of its findings. An April ReliaQuest briefing described claimed discoveries of longstanding FreeBSD and OpenBSD vulnerabilities, an autonomous four-vulnerability browser sandbox escape, and stronger Firefox exploit-development results than Opus 4.6. Anthropic withheld public access and offered restricted access through Project Glasswing to help participating organizations identify and fix software flaws. By September, Anthropic claimed its models had found 26,000 vulnerabilities, but only 2,000 had been reported to affected projects, with independent human review identified as the bottleneck. An October Pivot to AI report cited curl maintainers as saying Glasswing produced just one genuine new vulnerability, characterized as minor. Linux kernel maintainer Greg Kroah-Hartman described reports containing uninformative crashes, fabricated findings, and issues already discovered and fixed by others, while acknowledging four minor bugs credited to Anthropic. These accounts highlight substantial validation and triage costs rather than establishing that AI-assisted discovery has no value. For security leaders, the priority is to pair AI-assisted research with human verification, track validated and actionable findings rather than raw discovery totals, and maintain rapid vulnerability mitigation and detection workflows. The references raise concerns about faster exploit development but do not document an actual malicious campaign using Mythos.
Oct 9, 2026