Chinese AI Models DeepSeek and GLM Near OpenAI’s Top Cyber Model in Hacking Tests
Original headline: Free Chinese AI Models Are Getting Scary Good at Hacking
In a controlled experiment, AI models were tasked with breaching simulated networks, and the Chinese models DeepSeek and GLM demonstrated hacking capabilities close to those of OpenAI’s most advanced, yet unreleased, cyber model. The findings highlight that freely available Chinese AI systems are rapidly acquiring sophisticated offensive skills.
If free Chinese AI models attain advanced hacking abilities, existing cyber‑defense measures may become increasingly vulnerable.
Research angleWhat factors influence firms’ adoption of AI‑driven penetration testing tools for security investment decisions? — Technology Acceptance Model (TAM)
This story could become your next paper
HJITM works with researchers from study design through surveys, analysis and submission to international journals in generative AI and technology management. Bring us your research question.
This item is an AI-generated summary of reporting by WIRED. Please check the original article for accuracy. Copyright in the original article belongs to WIRED.