Research & Safety
Free Chinese AI Models Close in on OpenAI's Cyber Model in Hacking Test
In an experiment with simulated networks, DeepSeek and GLM performed not far behind OpenAI's most advanced cyber model, which is not public.
RESEARCH & SAFETY
Wired reports on an experiment in which AI models were asked to break into simulated networks. The hacking skills of DeepSeek and GLM were not far behind OpenAI's most advanced cyber model.
That OpenAI model is not public. The Chinese models are free to use.
Why it matters: Strong hacking skills in free models change the security baseline for anyone shipping software.
The result suggests that capable offensive tools are not limited to closed systems. The report does not say how the test was scored, so treat the comparison as a snapshot.
What builders should know
- Assume attackers can reach capable AI tools at no cost.
- Test your own systems with the same kinds of tools.
Sources: Wired