CrowdStrike Warns Claude Mythos Could Speed Zero-Day Decisions

Claude Mythos has pushed cybersecurity leaders into a new phase of AI planning after CrowdStrike saw demand surge around early model access.

Claude Mythos Access

Anthropic announced Project Glasswing and Claude Mythos Preview in Apr., drawing attention from security teams because the model is said to identify vulnerabilities across major browsers and operating systems.

Sentonas told Observer that interest jumped after CrowdStrike’s role in Glasswing became known, saying “the phones went crazy” as companies sought access that CrowdStrike could not provide.

“People wanted access to Mythos, which we obviously couldn’t provide. So we’ve been working around the clock advising organizations on how to think about it, and that hasn’t stopped. It’s still the number one topic we’re discussing,” he said.

CrowdStrike is involved in both Anthropic’s Project Glasswing and OpenAI’s Trusted Access for Cyber program, which give selected partners early access to frontier models for testing, internal use and feedback.

“We get access so we can test the models, use them internally, and understand how they might help customers,” Sentonas said. “Then we share feedback based on what we’re seeing.”

The company is also briefing CEOs, CIOs and CSOs on what Mythos can and cannot do, while trying to limit inflated expectations around a model described as “too dangerous to release.”

Also Read: Dogecoin Hollywood Director Gets 30 Months After Crypto And Stock Bets

Michael Sentonas Warning

The pressure increased after U.S. export controls forced Anthropic to suspend access to Mythos last month, though the Commerce Department lifted those restrictions this week.

CrowdStrike, with a market value of more than $194 billion and quarterly revenue of $1.4 billion, is a major voice in the debate because its Falcon platform uses AI to detect and respond to threats.

Sentonas said much of the work now involves separating practical risk from misinformation, including guidance on vulnerability scanning and cost management, as Mythos is priced at $50 per million output tokens. For CrowdStrike, Mythos has not produced major internal surprises because the company already uses models such as Opus for vulnerability scanning.

The larger concern is broader access, since attackers could eventually use comparable tools to find weaknesses and move faster than defenders can patch them.

“Imagine a world where 200 vulnerabilities are discovered every day. That’s not far away,” Sentonas said. “Now, imagine those vulnerabilities being exploited almost immediately. A CSO would have to decide, every day, which zero-day threat they need to mitigate.”

That warning follows CrowdStrike’s high-profile role in a global computer outage two years ago, a reminder that security vendors already operate in environments where software errors can produce worldwide consequences.

Read Next: Hackers Record 207 Crypto Hacks Expose Weak Spot In Digital Asset Security