Mozilla's Firefox 150 Ships with 151 Bugs Squashed, but Discord Crew Already Weaponized Leaked Mythos AI

Image: Bbc
Main Takeaway
Mozilla patched 151 hidden bugs before Firefox 150 shipped, but leaked Anthropic Mythos model is already being used by Discord hackers to find new exploits.
Jump to Key PointsSummary
What Mozilla Actually Fixed
Mozilla rolled Firefox 150 out this week carrying patches for 151 bugs that had never been reported. Anthropic’s yet-to-ship Mythos model, running inside a sealed red-team environment, produced the findings. Twenty-two of the issues reached the highest severity tier; the rest were medium- and low-risk crashes or memory-safety gaps. All fixes landed in the release branch before the browser shipped to the public.
How the Hunt Worked
Mozilla handed Anthropic a static export of the Firefox codebase plus a fuzzing harness that can replay every parser path a webpage can trigger. Mythos then spent roughly 48 hours generating test cases, crashing the browser thousands of times, and auto-triaging the root causes. Engineers at Mozilla reviewed each ticket, confirmed reproducibility, and rewrote the vulnerable code. The whole loop, AI finds, humans verify, patches land, took three weeks.
Why Mozilla Isn’t Worried About Skynet Yet
Firefox CTO Bobby Holley calls the exercise proof that defenders still hold the edge: the same tooling that can weaponize a flaw can also neuter it before release. Mozilla argues that scaling bug discovery simply moves the arms race earlier in the pipeline, giving maintainers time instead of attackers. The bigger risk, Holley warns, is developer disruption: teams must learn to triage AI-generated reports without drowning in noise.
The Breach That Wasn't Just a Breach
Anthropic's tight lid on Mythos slipped harder than first reported. A leaked debug token didn't just sit unused. Fortune AI reports a Discord group calling themselves "Mythos Underground" spent 11 days with live access before Anthropic noticed. They didn't just poke around. They built tooling. They found 14 fresh zero-days across Signal, Telegram, and three major VPN clients. They streamed the exploits live to 2,400 viewers. Anthropic's "no evidence of misuse" line looks quaint now.
Anthropic's Damage Control
The company revoked the token, sure. But the genie bolted. Fortune tracked down "cyberheretic," the Discord user who built the Mythos-to-Metasploit bridge. He claims the group shared exploit templates with 80+ members before the shutdown. Anthropic now admits the breach lasted longer than initially disclosed (11 days vs 3) and involved more users (hundreds vs "a tiny whitelist"). Their staged "Project Glasswing" rollout faces new scrutiny from AWS, Apple, and Microsoft partners who didn't sign up for this PR nightmare.
Mozilla's New Reality Check
Holley's "defenders hold the edge" argument just got stress-tested in public. Mozilla engineers are now racing to patch the same codebase against exploits that Mythos found after its escape. The Firefox team won't confirm how many of the Discord group's zero-days target their browser, but they're quietly pushing nightly builds with fixes that weren't in the original 150 release. The arms race didn't move earlier in the pipeline. It just went multiplayer.
Key Points
Discord hackers weaponized leaked Mythos for 11 days, finding 14 new zero-days beyond Mozilla's 151 bugs
Streaming exploits to 2,400 viewers, group shared templates with 80+ members before shutdown
Anthropic admits breach timeline tripled from 3 to 11 days, user base exploded from whitelist to hundreds
Mozilla quietly pushing new nightly patches as Discord group may have Firefox-specific exploits
Project Glasswing partners (AWS, Apple, Microsoft) re-evaluating deals after public weaponization
Questions Answered
11 days according to Fortune AI's investigation, not the 3 days Anthropic initially claimed.
They built a Metasploit bridge, found 14 new zero-days across Signal/Telegram/VPNs, and streamed exploits to 2,400 viewers.
Mozilla won't confirm specifics but is pushing new nightly builds with patches beyond the original 150 release.
AWS, Apple, Microsoft and other partners are reportedly re-evaluating their involvement after the public weaponization.
Source Reliability
47% of sources are highly trusted · Avg reliability: 75
Go deeper with Organic Intel
Simple AI systems for your life, work, and business. Each one includes copyable prompts, guides, and downloadable resources.
Explore Systems