Chinese Hackers Impersonate Top US AI Officials to Steal Policy Secrets

A new report from cybersecurity firm Proofpoint reveals that a China-aligned hacking group, tracked as TA419, has been impersonating high-profile U.S. artificial intelligence experts, including former White House officials, to steal sensitive email and login information from policy researchers at American and Japanese institutions. The sophisticated phishing campaign aimed at experts in AI regulation, export controls, and national AI strategy highlights a significant shift in Beijing's cyber espionage tactics, focusing not just on stealing technology, but on understanding the strategic direction of AI policy in rival nations. These cyber attackers, active since at least April 2025, used 'benign conversation starter emails' to lure their targets, posing as figures like Lynne Parker, a former White House technology official, and Heidi Crebo-Rediker, a former State Department economist, inviting them to fictitious 'AI Policy Advisory Committee' or Senate reports on AI. Once targets engaged, the hackers would then direct them to fake OneDrive pages or use advanced tools like 'Frameless BitB' to steal Microsoft login credentials. This aggressive intelligence gathering comes amidst a heated U.S.-China AI race, with Washington already expressing concerns about China's industrial-scale efforts to siphon American AI capabilities. The revelations from Proofpoint underscore the escalating digital tug-of-war for AI dominance, prompting increased vigilance from cybersecurity and national security agencies. While Beijing routinely denies state-backed hacking, these incidents will likely fuel further diplomatic tensions and accelerate calls for stronger defenses around critical AI research and policy-making. Expect more targeted attacks as the global competition for AI leadership intensifies, forcing universities, think tanks, and government bodies to fortify their digital perimeters against increasingly clever state-sponsored adversaries.