TL;DR
Anthropic is expanding Project Glasswing from an initial group of about 50 partners to about 150 new organizations across more than 15 countries. The move follows reports that early partners using Claude Mythos Preview found more than 10,000 high- or critical-severity security flaws, shifting the work from finding bugs to verifying, disclosing, patching and deploying fixes.
Anthropic is expanding Project Glasswing to roughly 150 new organizations after an initial group of about 50 partners using Claude Mythos Preview found more than 10,000 high- or critical-severity security flaws, according to source material describing the program. The development matters because the program is now aimed less at finding vulnerable code and more at the slower work of confirming, disclosing, patching and deploying fixes across software used in critical infrastructure.
Project Glasswing is Anthropic’s collaborative effort to secure widely used software. In early April, about 50 initial partners were given access to Claude Mythos Preview and began scanning their codebases for vulnerabilities. The reported result was more than 10,000 high- or critical-severity findings.
The new group includes organizations in more than 15 countries and expands into sectors that were underrepresented in the first cohort, including power, water, healthcare, communications, hardware and software vendors. According to the source material, many of the partners serve critical infrastructure, and a successful attack on some partners’ codebases could affect more than 100 million people.
Anthropic is also moving the program’s focus downstream. Partners are using Mythos-class models not only to scan code, but also to write patches, run pre-release checks, support penetration testing and rebuild legacy code in memory-safe languages. Anthropic is also said to be discussing ways to scale review and patching of open-source vulnerabilities while sharing disclosure practices meant to help maintainers handle AI-found reports.
The bottleneck moved — from finding flaws to fixing them
50 partners found 10,000+ critical vulnerabilities in weeks. So the constraint is no longer detection — it’s verify, disclose, patch, deploy. Anthropic is expanding Project Glasswing to ~150 organizations, and pivoting its weight toward the new chokepoint.
From 50 partners to ~150 — aimed at the leverage points
Not just more headcount. The new group reaches sectors the first cohort underrepresented, and leans toward vendors whose code sits under thousands of downstream systems.
each must meet Anthropic’s security requirements first
cybersecurity vulnerability scanning tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Amazon Product B07SLBTLF3
As an affiliate, we earn on qualifying purchases.
Finding used to be the hard part
For the whole history of the field, detection was the scarce, skilled work — the chokepoint. A model that surfaces 10,000 critical flaws in weeks inverts that. Toggle before/after and watch the bottleneck move.
The defensive pipeline — where the constraint sits
Same five stages. The chokepoint slides downstream.
Cute-Patch It Works on My Machine Meme Embroidered Iron on sew on Patch Funny Emblem Programmer Humor
Size: 3 inches tall
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Amazon Product B0GFXPFLY6
As an affiliate, we earn on qualifying purchases.
AI redeployed downstream — and pushed beyond the cohort
Glasswing is consciously shifting its weight from finding toward disclosing, fixing & deploying. The same model helps at the new bottleneck.
Defensive tasks Mythos-class models now take on
Beyond scanning — the work that actually closes the gap.
Writing patches
Partners use the model to fix what it finds — not just flag it.
Pre-release checks
Preventing vulnerabilities from appearing in the first place.
Penetration testing
Simulating attacks to see how a flaw might be exploited.
Rebuilding in memory-safe languages
Attacking whole vulnerability classes at the root.
Claude Security
Uses public frontier models like Claude Opus 4.8 to scan codebases & suggest patches.
The Glasswing tooling
The vuln-finding tools, to trusted security teams — so partners’ methods replicate widely.
code vulnerability detection software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Why the urgency is named, not gestured at
The program’s tempo is the tempo of a race against diffusion. Anthropic puts a number on the deadline.
Within 6–12 months, many other labs will have Mythos-class models — and could release them without safeguards.
In that world, cyberattacks could occur much more often, and in much more unpredictable forms. The strategic theory of the whole program: build the defensive head start now, while the capability is still scarce and gated — so when it’s cheap and everywhere, defenders already stand on higher ground.
Capability is scarce & gated
Mythos-class power sits with vetted Glasswing partners under Anthropic’s requirements.
Capability goes ambient
Other labs ship Mythos-class models — possibly ungoverned. The window to prepare closes.
CompTIA CySA+ Certification Kit: Exam CS0-003 2025-2026: A Complete Cybersecurity Analyst Study System for Mastering Threat Detection, Incident Response and Security Monitoring With 1000 practice
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Read it with its difficulties in view
Several are real — some Anthropic states outright, some inherent to the situation. None cancels the core, but all deserve to be held.
Dual use — and the safeguards don’t exist yet
The same capability that finds-and-patches can find-and-exploit. Anthropic says general release needs safeguards that it, and to its knowledge all other developers, have yet to develop. The caution is the clearest evidence of the power.
Gated, even as the logic demands breadth
Advanced defensive capability is allocated by one company’s selection — yet the announcement’s own case is that hundreds of thousands will need access. “Must be gated for safety” sits in tension with “must be widespread to work.”
Not a neutral observer
A frontier lab is at once warning of the danger, helping constitute it, and selling the response (Claude Security, the tooling, the Cyber Verification Program). The warning isn’t wrong — but the commercial frame is worth holding alongside the public-interest one.
Toward a permanent advantage for defenders
Cybersecurity has long been asymmetric in the attacker’s favor — defenders close every hole, attackers need one. The north star is to flip that.
More essential infrastructure
Plus critical-OSS maintainers & safety testers, US & overseas.
Cyber Verification Program
Mythos-class capability for specific cyberdefense tasks — breadth without waiting on full-release safeguards.
Make all software secure
And help the industry adjust how AI changes the core assumptions of cybersecurity.
Reading it in proportion
- The core is hard to argue with: AI made finding cheap & abundant; the bottleneck genuinely moved to patching & deployment; redirecting effort there is sane.
- The caveats sit alongside, not against: one company’s program, one company’s gate, a timeline & products that company has reason to advance — and admittedly-missing release safeguards.
- Hold both halves: the danger is plausible and the 10,000 flaws are real; the response is reasonable and commercially convenient; the aspiration is worthy and unproven.
Why It Matters
The expansion points to a shift in the cybersecurity problem created by stronger AI systems. If automated tools can surface thousands of serious flaws in weeks, the limiting factor becomes the human and organizational pipeline that follows: verifying whether findings are real, coordinating responsible disclosure, building fixes, testing them and getting updates into production systems.
That matters for readers because many widely used systems depend on software maintained by vendors, infrastructure operators and open-source projects. A large increase in vulnerability discovery can improve defense only if fixes reach users before attackers can exploit the same weaknesses.
Background
The first phase of Project Glasswing gave about 50 trusted partners access to Claude Mythos Preview under Anthropic’s security requirements. The reported findings from that phase reframed the program’s purpose: detection appeared to be less scarce than the capacity to act on the findings.
The expansion to about 150 new organizations is framed as a way to reach higher-leverage codebases, including vendors whose software is used by many downstream organizations and governments. Anthropic has also released Claude Security for the general market, using public frontier models such as Claude Opus 4.8 to scan codebases and suggest patches, while the Glasswing tooling remains available on request to trusted security teams.
What Remains Unclear
Several details remain unclear from the provided source material. Anthropic has not been quoted here naming the full partner list, the exact number of confirmed exploitable vulnerabilities, how many flaws have already been patched, or how the findings are distributed by sector, severity or software type. It is also unclear how many open-source maintainers will receive reports, how those reports will be prioritized, and what safeguards govern broader access to the Glasswing tooling.
What’s Next
The next phase is expected to center on whether partners can turn AI-generated findings into verified fixes that are disclosed, patched and deployed at scale. Key markers will include patch rates, disclosure timelines, adoption by critical infrastructure operators and any public reporting from Anthropic on how many vulnerabilities were confirmed and resolved.
Key Questions
What is Project Glasswing?
Project Glasswing is Anthropic’s collaborative program to help trusted partners scan important software for security flaws and use AI systems to support remediation work.
What is the main news development?
Anthropic is expanding the program from an initial group of about 50 partners to roughly 150 new organizations after early participants reported more than 10,000 high- or critical-severity findings.
Why is the expansion focused on fixing, not just finding?
The reported volume of findings means discovery is no longer the only constraint. The hard work now includes confirming findings, notifying affected parties, writing fixes, testing them and getting updates deployed.
Which sectors are included in the expanded group?
The source material lists power, water, healthcare, communications, hardware and vendors, with many participants tied to critical infrastructure or software used by many downstream organizations.
What remains unknown?
The source material does not provide the full partner list, patch completion rates, a detailed breakdown of the vulnerabilities, or proof that all reported findings are exploitable. Those details are still developing.
Source: Thorsten Meyer AI
