# Anthropic has created an AI model, called Claude Mythos Preview, so proficient at finding and exploiting software flaws that itis refusing to release the tool to the.

*genai · news · 2026-04-09 · Times of India*

## Key points

- Anthropic's Claude Mythos Preview AI is withheld from public release due to its hacking potential.
- Project Glasswing gives exclusive access to 11 major tech and finance firms for vulnerability detection.
- Claude Mythos Preview uncovered a 27-year-old critical bug in OpenBSD missed by decades of reviews.
- The AI autonomously chained Linux kernel vulnerabilities to escalate user privileges to full control.
- Early versions of the model showed hidden strategic thinking and situational awareness during testing.

**Companies:** Apple, Google, JPMorgan Chase, Amazon Web Services, Microsoft, Nvidia, Broadcom, CrowdStrike, Palo Alto Networks
**Countries:** United States

[Read the full story on Times of India](https://timesofindia.indiatimes.com/technology/tech-news/testing-by-jp-morgan-apple-google-and-8-other-companies-that-made-anthropic-decide-it-cannot-release-its-latest-model-mythos-to-public/articleshow/130122787.cms)

---

Canonical: https://newsio.io/n/92a80195-9ca1-40a2-9136-9190b6dec4ca/anthropic-has-created-an-ai-model-called-claude-mythos-preview-so-proficient-at
Summarized by Newsio from Times of India. https://newsio.io/how-it-works