# The software bug was capable of crashing an operating system used by firewalls, servers and network appliances.

*fintech · news · 2026-04-14 · Hindustan Times*

## Key points

- Anthropic's AI model Mythos found a 27-year-old critical bug in OpenBSD previously undetected.
- Mythos and similar AI tools are uncovering software bugs at unprecedented rates, overwhelming developers.
- AI models can now write exploit code for discovered vulnerabilities, drastically reducing required hacker skill.
- The average time from bug disclosure to exploitation has dropped from 847 days to less than one day.
- Anthropic is not releasing Mythos publicly due to safety concerns and is working with about 50 organizations.

**Companies:** OpenAI, Google, Anthropic, Stripe, HackerOne, Sysdig, Fly.io, Numeric
**Countries:** United States

[Read the full story on Hindustan Times](https://www.hindustantimes.com/technology/ai-is-finding-bugs-that-hackers-can-exploit-get-ready-for-bugmageddon-101776137774550.html)

---

Canonical: https://newsio.io/n/68aba0a8-d97f-4eba-ad05-87cb44e76441/the-software-bug-was-capable-of-crashing-an-operating-system-used-by-firewalls-s
Summarized by Newsio from Hindustan Times. https://newsio.io/how-it-works