Chinese researchers adapt Meta's Llama model for military AI tool ChatBIT

Chinese researchers have adapted Meta's open-source Llama model to create a military-focused AI tool called ChatBIT. This development raises security concerns, as it marks the first known instance of a national military repurposing Llama for defense purposes. ChatBIT was developed by researchers linked to the People's Liberation Army and is designed to assist in military operations. It reportedly performs at about 90% of GPT-4's capacity, although the methods used to measure this performance are unclear. Meta has distanced itself from the military use of Llama, emphasizing that such applications violate its licensing terms. The situation has prompted discussions in the U.S. about stricter export controls to prevent foreign military access to open-source technologies.