TL;DR
Anthropic is expanding Project Glasswing from an initial group of about 50 partners to roughly 150 new organizations. The move follows reports that early partners found more than 10,000 high- or critical-severity vulnerabilities, shifting attention from detection to verification, disclosure, patching and deployment.
Anthropic is expanding Project Glasswing to roughly 150 additional organizations after an initial group of about 50 partners used Claude Mythos Preview to find more than 10,000 high- or critical-severity software vulnerabilities, a result that shifts the program’s focus from finding flaws to helping security teams verify, disclose, patch and deploy fixes.
The expansion brings Project Glasswing beyond its early April cohort, according to the source material. Anthropic’s program gives vetted partners access to Claude Mythos Preview and related tooling for security work on major software systems.
The new partner group is expected to span more than 15 countries and add representation from power, water, healthcare, communications, hardware and vendor ecosystems. The source material says the selected organizations are tied to critical infrastructure or widely used software, meaning a successful attack on some partners’ code could affect large downstream populations.
Anthropic is also widening the work beyond vulnerability scanning. According to the source material, partners are using Mythos-class models to write patches, run pre-release checks, perform penetration testing and rebuild some legacy code in memory-safe languages. Anthropic is also said to be in talks to scale review and patching of open-source vulnerabilities and to share disclosure practices for maintainers.
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.
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.
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.
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.
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 development matters because it points to a new pressure point in AI-assisted cybersecurity. If advanced models can identify severe flaws at high volume, the limiting factor becomes whether human teams and software maintainers can confirm the findings, coordinate disclosure, ship patches and get fixes deployed before attackers can exploit the same classes of weaknesses.
That burden is especially heavy for open-source projects and infrastructure vendors whose code may sit inside many other systems. A large influx of AI-found vulnerabilities could help defenders, but it could also overwhelm maintainers if reports are poorly validated or lack usable remediation details.
Background
Project Glasswing began with roughly 50 initial partners in early April. Those partners were given access to Claude Mythos Preview and began scanning codebases for vulnerabilities. The reported discovery of more than 10,000 high- or critical-severity flaws is the central data point behind the expansion.
The source material frames the program as a race against diffusion of similar model capabilities. Anthropic is described as estimating that within six to 12 months, many other labs may have models in the same class and could release them with fewer safeguards. That timeline is Anthropic’s claim, not an independently verified forecast.
What Remains Unclear
Several details remain unclear. The source material does not name the new partner organizations, list the affected products, verify how many of the reported vulnerabilities have been independently confirmed, or say how many have already been patched. It is also unclear how Anthropic will measure the program’s success beyond vulnerability counts, or how disclosure will be managed when flaws affect open-source maintainers with limited capacity.
What’s Next
The next test for Project Glasswing is whether the expanded partner group can turn large-scale detection into completed remediation. That means validated reports, coordinated disclosure, shipped patches and deployed fixes, especially in software used by infrastructure operators, governments and downstream vendors.
Key Questions
What is Project Glasswing?
Project Glasswing is Anthropic’s collaborative effort to help secure major software systems using Claude Mythos Preview and related security tooling made available to vetted partners.
What is new in the expansion?
Anthropic is extending the program from an initial group of about 50 partners to roughly 150 additional organizations, with a focus on sectors and vendors tied to critical infrastructure and widely used code.
What did the first partners find?
According to the source material, the initial partners found more than 10,000 high- or critical-severity vulnerabilities after receiving access in early April.
Why is the focus shifting from scanning to fixing?
Large-scale detection can create a backlog. The source material says the hard part is now verification, disclosure, patching and deployment, because flaws do not reduce risk until fixes reach affected systems.
What remains unknown?
The identities of many partners, the affected products, the validation status of the reported flaws and the number of completed patches have not been provided in the source material.
Source: Thorsten Meyer AI
