CIA to Deploy Generative AI 'Co-Workers' to Counter Foreign Espionage

The US Central Intelligence Agency is moving to embed "AI co-workers" across its entire analytic infrastructure over the next two years. Deputy Director Michael Ellis stated that these tools will assist intelligence officers in drafting key judgments, testing analytical conclusions, and identifying patterns within global intelligence data to better anticipate hostile moves by foreign adversaries. This initiative is driven by a perceived narrowing of the technological innovation gap between the United States and China. Ellis noted that while the US once held a significant lead, that advantage has diminished, making the adoption of advanced AI critical for maintaining national security. The agency's push for internal capabilities follows a public dispute between the Department of Defense and AI firm Anthropic. Following a March order from President Donald Trump, federal agencies ceased using Anthropic's technology after the firm restricted its use for mass surveillance and autonomous weapons, leading the DoD to designate the company a supply chain risk. The CIA has already piloted approximately 300 AI projects focused on language translation and large-scale data processing. While the agency has produced its first AI-generated intelligence report, Ellis emphasized that human officers will remain the final decision-makers for critical intelligence judgments. Beyond generative AI, the agency is increasingly utilizing blockchain data for counterintelligence operations. Ellis highlighted Bitcoin and cryptocurrency as critical matters of national security in the broader technological competition with global adversaries.