Brief by Shorts91 Newsdesk / 09:10am on 30 Sep 2026,Wednesday Tech Today
Chinese AI developer Moonshot is reviewing its Kimi models after security researchers found they could be tricked into discussing biological weapons and assassinations. Mindgard, an AI security firm, said it discovered in July that Kimi K2.6 and K3 Swarm could bypass safety restrictions through a process known as “jailbreaking”. The researchers also warned that a compromised model could potentially help hackers run code and launch cyberattacks. However, Mindgard said it had not verified whether the biological weapon instructions provided by the models would work. Moonshot said it welcomed external feedback and was discussing the findings with Mindgard. (PC: BBC)
