# Free Chinese AI Models Are Getting Scary Good at Hacking

- Source: Wired · AI
- Published: 2026-11-06T18:43:43.000Z
- Bad Influence: https://badinfluence.news/items/j9se7i4cduzjz1z1es9axtukn
- Original: https://www.wired.com/story/free-chinese-ai-models-are-getting-scary-good-at-hacking/

## Summary

Vals, an AI capabilities measurement company, tested models on multi-step hacking tasks across two simulated corporate networks. OpenAI's Daybreak, available only to trusted partners, completed 81 percent of tasks on average, while DeepSeek completed 73 percent. Vals says an attack costs $17 to $46 in tokens with GPT-6-Sol compared with $2 to $3 with DeepSeek.