Cybercriminals associated with the Chinese government have carried out a sophisticated scheme, posing as Anthropic employees to gather confidential information about AI development in the U.S. This information comes from a report by the Silicon Valley cybersecurity firm Proofpoint, which outlines the activities of a hacking group referred to as “TA419.”
TA419 targeted AI policy specialists at U.S. think tanks, universities, and law firms, masquerading as industry insiders, including a former White House official. The hackers devised seemingly innocuous introductory emails, inviting their targets to participate in a fictitious “AI Policy Advisory Committee” or to contribute to a phony Senate Committee on Foreign Relations report. Their deceptive strategies intensified after initial exchanges, culminating in the delivery of documents infected with malware intended to compromise the victims’ cloud accounts.
For example, one email featured the subject line “Request for Feedback on Military Integration of Claude,” aimed at an individual affiliated with a think tank. “We are confident,” Mark Kelly, a researcher at Proofpoint, stated in an interview with CNN, asserting that the group is a “Chinese government-aligned threat actor.” He cited the consistent targeting of interests aligned with the Chinese government, observed infrastructure, and technical artifacts, along with corroboration from industry partners as indications of their alignment.
Among those impersonated was Lynne Edwards Parker, who served as Principal Deputy Director of the White House Office of Science and Technology Policy under Joe Biden. She told CNN that in addition to concern for colleagues, the suspected espionage scheme left her with a sense of “sadness that relationships I’ve built over my career were being exploited by bad actors to deceive others.”
This incident takes place during a notably tense time for AI development in the U.S. Public opinion favors the introduction of regulations on AI technologies, a perspective shared by several leading AI executives, including Dario Amodei of Anthropic. Their worries focus on the possibility of uncontrolled AI leading to significant disruptions, whether through security breaches or catastrophic infrastructure failures.
In spite of these calls for regulation, the White House, alongside industry figures such as Meta’s Mark Zuckerberg and Nvidia’s Jensen Huang, has resisted moves to impose stricter oversight. Recently, the administration hosted a summit on September 29, 2026, where tech executives signed a document that President Donald Trump described as a “morally-binding” agreement aimed at enabling “tremendous self-policing.”
Trump has made his position on the competitive landscape of AI clear, declaring, “Whoever wins AI, WINS!” in a post on Truth Social. This statement reflects the prevailing argument among those against regulation: that the U.S. cannot afford to forfeit its lead in the international AI arms race to China.
Proofpoint’s report concludes that TA419 “will likely continue targeting think tanks and policy experts working on technologies, and in geographies, of particular interest to the Chinese government.” The firm added that “these campaigns will likely also continue spoofing the identities of real subject-matter experts.”
