Cardano

Open Source

Frontier

Security

Mythos

Anthropic says it intends to give the public access to <a href="https://yellow.com/news/anthropic-mythos-ai-10000-critical-bugs" rel="noopener noreferrer nofollow">Mythos, its vulnerability-hunting AI</a> model, but only once it builds safeguards that do not yet exist.
<blockquote>
 <h2 id="key-points">Key Points:</h2>
 <ul>
 <li>Anthropic plans to release Mythos-class models broadly after first expanding access to U.S. and allied governments.</li>
 <li>The company admits no firm, including itself, has built safeguards strong enough to stop misuse.</li>
 <li>Mythos has flagged more than 23,000 issues across 1,000 open-source projects, including 6,202 high or critical flaws.</li>
 </ul>
</blockquote>
<h2 id="anthropic-mythos-release">Anthropic Mythos Release</h2>
Anthropic <a href="https://www.anthropic.com/research/glasswing-initial-update" rel="noopener noreferrer nofollow">confirmed</a> the plan in an update to Project Glasswing, its limited-access security program, and a separate report flagged the timeline as uncertain.
The company said it would first work with U.S. and allied governments to widen the program. A broader release of "Mythos-class models" would follow in the near future.
Anthropic was blunt about the risk. It <a href="https://www.theregister.com/security/2026/05/25/anthropic-to-release-mythos-class-models-to-the-public/5245596" rel="noopener noreferrer nofollow">stated</a> that no company, itself included, has built safeguards strong enough to keep the model from being misused and causing severe harm.
Even so, the company expects similar tools to spread fast, predicting Mythos-level models will reach wide availability within six to 12 months.
Mythos<a href="https://www.nbcnews.com/tech/security/anthropic-project-glasswing-mythos-preview-claude-gets-limited-release-rcna267234" rel="noopener noreferrer nofollow">debuted</a> in April. Anthropic said it generated working exploits 72.4 percent of the time in testing, against nearly zero percent for an earlier Claude model.
Also Read:<a href="https://yellow.com/news/cisco-research-frontier-ai-models-multi-turn-attacks" rel="noopener noreferrer nofollow">Cisco Research Shows Frontier AI Models Failing Under Multi-Turn Attacks</a>
<h2 id="mythos-vulnerability-findings">Mythos Vulnerability Findings</h2>
Since its debut, the model has scanned more than 1,000 open-source projects and <a href="https://yellow.com/news/anthropic-fsb-mythos-83-percent-exploit" rel="noopener noreferrer nofollow">surfaced 23,019 issues</a>, of which 6,202 were rated high or critical severity.
One finding stood out. Mythos uncovered a flaw in the wolfSSL cryptography library, used by billions of devices, that could have let attackers forge certificates and impersonate banks or email providers. It has since been patched.
The flood of reports has strained the people meant to fix them. Open-source maintainers have asked Anthropic to slow disclosures, saying the volume outpaces their capacity.
Researchers see a deeper imbalance. Anthropic argues that finding bugs is now far easier than fixing them, and the company <a href="https://www.helpnetsecurity.com/2026/05/26/anthropic-project-glasswing-update/" rel="noopener noreferrer nofollow">partnered</a> with the Open Source Security Foundation's Alpha-Omega project to help maintainers triage the backlog.
The Claude Mythos system card predicts AI will eventually favor defenders, though Anthropic concedes attackers may hold the edge for now.
When <a href="https://yellow.com/news/anthropic-mythos-coinbase-interest" rel="noopener noreferrer nofollow">Mythos was first revealed</a>, Anthropic gave more than 50 organizations, including Apple, Microsoft and Google, access alongside roughly $100 million in usage credits, withholding the model from the public over its capacity to weaponize software flaws.
Read Next:<a href="https://yellow.com/news/cardano-whales-seize-675-ada-supply" rel="noopener noreferrer nofollow">Cardano Whales Seize 67.5% Of ADA Supply, A Six-Year High</a>

Anthropic Will Open Claude Mythos To The Public Once Safeguards Catch Up, Maybe In 12 Months

Table of Contents

Yellow News

Other articles published on May 28, 2026

Join our free newsletter for daily crypto updates!