The Pentagon is developing alternatives to Anthropic, report says

Following a significant breakdown in relations, the Pentagon is pivoting away from its collaboration with Anthropic and is instead developing alternative artificial intelligence tools. Cameron Stanley, the Pentagon's chief digital and AI officer, revealed in a Bloomberg discussion that the Department of Defense (DoD) is currently engaged in creating various large language models (LLMs) tailored for government use. "The Department is actively pursuing multiple LLMs into the appropriate government-owned environments," Stanley noted, emphasizing that engineering efforts are underway to make these models operational in the near future. The rift between Anthropic and the Pentagon emerged after discussions broke down over a $200 million contract. A key sticking point was the level of unrestricted access the military could have to Anthropic's AI systems. While Anthropic aimed to include provisions to prevent the use of its technology for mass surveillance or autonomous weaponry, the Pentagon remained firm in its stance. In the wake of this disagreement, OpenAI has stepped in, securing its own partnership with the Pentagon. Additionally, the Department of Defense has formed an agreement with Elon Musk's xAI to utilize Grok in classified operations. This shift underscores the Pentagon's strategy to move away from Anthropic's technology, especially after Defense Secretary Pete Hegseth labeled Anthropic as a supply chain risk—a designation typically associated with foreign adversaries, which restricts Pentagon contractors from engaging with them. Despite some speculation about the potential for reconciliation, the Pentagon appears set on advancing without Anthropic. The company is currently contesting the supply chain risk designation in court, marking a tumultuous chapter in its relationship with the DoD.