2026-08-25 · Source: Insurance Journal
Summary in 3 Points • Chinese hackers are using AI models like DeepSeek to enhance cyberattacks • TeamT5 reports a significant increase in attacks by state-affiliated groups • Hackers prefer DeepSeek for its low cost and customisation capabilities --- Chinese hackers have significantly increased their cyberattacks by integrating AI models such as DeepSeek into their operations. According to Taiwanese research firm TeamT5, state-affiliated cyber groups have more than doubled their attacks by using AI to handle routine tasks and develop sophisticated malicious software. DeepSeek is particularly favoured for its high performance, customisation options, and low operational costs. Despite the availability of more powerful models like Moonshot's Kimi K3, hackers find DeepSeek's lax cybersecurity barriers more appealing. The AI has been used in various stages of attacks, including reconnaissance and exploiting vulnerabilities. Additionally, American AI models like ChatGPT have been employed in specific instances to assist in attacks.
The increasing use of AI by Chinese hackers to enhance cyberattacks could have direct consequences for the London insurance market. Underwriters may need to reassess cyber risk models to account for the heightened threat posed by AI-driven attacks. Policy wording might require updates to address potential coverage for damages caused by AI-enhanced cyber incidents. Furthermore, pricing strategies could be impacted as insurers evaluate the increased exposure to sophisticated cyber threats. The market may also see a rise in claims related to AI-driven cyberattacks, necessitating a review of claims handling processes and potential collaboration with cybersecurity experts.