Gate News message, April 22 — Anthropic’s latest AI model, Claude Mythos, identified 271 security vulnerabilities in Mozilla Firefox during internal testing, with all flaws remedied this week, Mozilla disclosed. In contrast, a previous version of Anthropic’s model had detected only 22 security-sensitive vulnerabilities.
Mozilla stated that all discovered vulnerabilities fell within the scope of findings by top human security researchers. Claude Mythos, launched in March 2026, is Anthropic’s most powerful model for reasoning, coding, and cybersecurity applications. The model is currently available to vetted partners including Amazon, Apple, and Microsoft through the “Project Glasswing” program.
Related News
Amazon adds more for Anthropic with $25 billion: 5GW of computing power, with a $100 billion-plus AWS commitment
Axios exclusive: The U.S. NSA bypassed the Pentagon blacklist to use Anthropic Mythos, and Dario Amodei urgently met with the White House to negotiate
Claude Opus 4.7 hides the price increase: a new tokenizer makes the same text use 37–47% more tokens, while the fee rate stays the same but the bill gets more expensive