> ## Content Index
> Fetch the complete content index at: https://nordisk-ai.ghost.io/llms.txt
> Use this file to discover other available public pages before exploring further.

# China Develops Military AI Model Based on Meta's Llama
- URL: https://nordisk-ai.ghost.io/china-develops-military-ai-model-based-on-meta-s-llama/
- Published: 2024-12-05T07:41:08.000Z
- Updated: 2026-08-27T17:20:53.000Z
- Description: China is adapting Meta's open-source AI model for military purposes. Is this a step forward for their military innovation, and what does it mean for future AI development and global security risks?
- Author: Anders Hagen
- Tags: Application Areas, AI Models, The People's Liberation Army, Generative AI, China, Llama, Meta, Military, military applications, Pentagon, security risks, #Migrated-1787841634512, #wp, #wp-post, #Import 2026-08-27 14:41, #Import 2026-09-07 13:41

According to recently published academic papers, Chinese research institutions with ties to the People's Liberation Army (PLA) have used an earlier version of Meta's open AI model Llama to build an AI model tailored for military purposes. The model, which the researchers call "ChatBIT," is designed to gather and process intelligence and support decision-making in military operations. The study, carried out by researchers at the PLA's leading research center, the Academy of Military Science (AMS), shows that ChatBIT outperforms several other AI models and performs close to OpenAI's ChatGPT-4.

#### Meta: PLA's Use Is Unauthorized

Meta has expressed its displeasure over the PLA's use of Llama, pointing out that it violates its terms of use, which prohibit military, nuclear, and espionage-related applications. A Meta spokesperson, Molly Montgomery, confirmed the company is actively working to prevent unauthorized use of its models. Despite this, the open availability of the Llama model means Meta has limited ability to strictly control who uses the technology.

#### Use Cases and Future Plans for ChatBIT

ChatBIT is optimized for military conversation and question-answering tasks and is planned for future use in strategic planning, simulation training, and command decisions. According to the AMS researchers, the model was built using just 100,000 military dialogues, a modest amount of data compared to other LLMs trained on far larger datasets.

#### The Pentagon and Security Risks Around Open AI Models

The US government has recently increased its focus on the security risks of open AI models. In October 2023, US President Joe Biden signed an executive order addressing AI development, emphasizing that innovation must be balanced with safety measures. The Pentagon said it's closely tracking competitors' AI capabilities and monitoring potential security risks from open models like Llama.

#### China Uses Open AI Models in Both Military and Security Contexts

In addition to ChatBIT, Chinese researchers have also used the Llama model in airborne electronic warfare as well as for policing and security surveillance. PLA Daily, a state-run newspaper, has also published commentary on how AI could accelerate the development of military weapons and streamline combat training. Despite restrictions from American companies, it's clear that China continues to use open AI models to strengthen its technological and military influence.