A Chinese AI company has launched an internal investigation after an AI researcher discovered one of the company's models could be made to provide instructions for developing biological weapons and carrying out assassinations. 

Researcher Peter Garrigan told Fox News that the company's Kimi model could be used to plan terrorist attacks using real-time data, create sarin nerve gas, develop malware and destroy aircraft. 

"We've also seen these problems within the U.S. models as well. It's a fundamental flaw in the technology," Garrigan claimed. 

Moonshot AI is investigating Garrigan's findings, and he is coordinating with the company in that investigation.