State-sponsored hackers from North Korea and China have shown a keen interest in using artificial intelligence (AI) to identify new cybersecurity vulnerabilities, as per a report by Google. The tech giant’s threat intelligence group highlighted significant activity from threat groups linked to China and North Korea in this regard. Specifically, North Korean hackers from group APT45 were mentioned for their advanced use of AI in cybersecurity research.
The report detailed how North Korean hackers employed AI to analyze potential cybersecurity weaknesses through repetitive prompts. Additionally, Google’s AI systems successfully detected and prevented a criminal group’s attempt to exploit a “zero-day exploit” for mass attacks. This marks the first instance where Google has observed malicious actors using AI to identify and exploit vulnerabilities on a large scale.
Amid growing global cybersecurity concerns, Google’s report coincides with the introduction of Claude Mythos, a new AI model by U.S. startup Anthropic specializing in identifying software security flaws. Anthropic has restricted access to this model, providing it only to selected entities for security testing purposes. The report comes at a time when South Korean stock markets experienced a decline due to investor profit-taking and uncertainties surrounding U.S.-Iran peace talks.
