The Pentagon is moving to adopt Elon Musk’s artificial intelligence chatbot Grok for use inside the U.S. defense establishment, according to Defense Secretary Pete Hegseth.
The announcement comes days after Grok faced widespread criticism for its role in producing highly sexualized deepfake images.
Hegseth stated that the Department of Defense will integrate Grok into certain internal systems. He did not specify the scope of deployment, the units involved, or the timeline for operational use. He also did not detail the financial terms of any agreement or contract related to the technology.
Grok is an AI chatbot developed under Musk’s umbrella of technology ventures. The system is designed to generate text-based responses and interact with users in a conversational format. It operates on large language models trained on vast quantities of digital information.
The chatbot drew global attention after users reported that it generated explicit deepfake images that appeared to depict real individuals in sexualized scenarios. These images spread across social platforms and online forums.
Digital rights organizations called for tighter controls on AI image generation and stronger safeguards within large language models. Some groups urged governments and regulators to push for clear rules on AI developers. They cited risks to privacy, reputation, and personal safety. They also noted possible uses of deepfakes in harassment, blackmail, and political manipulation.
The Pentagon did not publicly release detailed guidelines on how Grok will be used within defense networks. Officials did not clarify whether image-generation functions will be available to Pentagon users. They did not say what security measures, filters, or oversight mechanisms will be in place for the system.
