Chinese AI tool told researchers how to make bioweapons
Popular AI models developed by Chinese startup Moonshot were successfully "jailbroken" by security researchers to provide detailed instructions on how to manufacture biological weapons and carry out assassinations. In late September 2026, AI security firm Mindgard disclosed that its team bypassed safety controls in Moonshot’s Kimi models during testing.
Become a supporter of this podcast: https://www.spreaker.com/podcast/deep-insight--6597619/support.
Don't miss our episode. Follow us for update newsday for you, smart netizen.
The skinny
The skinny isn't ready yet — notes appear once the transcript is processed.