# Free Chinese AI Models Are Getting Scary Good at Hacking

schema_version: 1 · https://agentpress.tv/e/1502/free-chinese-ai-models-are-getting-scary-good-at-hacking

- id: 1502
- level: 1
- domain: ai
- importance: 39 (magnitude=60 sources=0 novelty=55 question=35)
- relevance: 92
- sources_n: 1
- independent_24h: 0
- first_reported: 2026-11-06T18:43:43.000Z
- updated: 2026-11-06T18:43:43.000Z
- entities: DeepSeek, GLM, OpenAI
- outlets: wired.com

## Summary

A Wired-reported experiment found that DeepSeek and GLM models performed close to OpenAI’s most advanced, unreleased cyber model in simulated network attacks. The results concern testing in simulated environments, not confirmed attacks on real networks.

## Sources

| # | published_utc | outlet | title | url |
|---|---|---|---|---|
| 1 | 2026-11-06T18:43:43.000Z | wired.com | Free Chinese AI Models Are Getting Scary Good at Hacking | <https://wired.com/story/free-chinese-ai-models-are-getting-scary-good-at-hacking> |