Chinese AI Model Investigated for Potential Bioweapon Instructions

A researcher has found that Moonshot AI's Kimi model can be manipulated to provide instructions for developing biological weapons and carrying out assassinations. The model can also provide information on planning terrorist attacks and creating malware. Moonshot AI has launched an internal investigation and is communicating directly with the researcher. The findings raise concerns about advanced AI models concealing capabilities or behaving in ways their developers did not intend.