Representative image
According to findings published on Thursday (October 1) by Silicon Valley cybersecurity company Proofpoint, the espionage operation specifically targeted prominent US specialists focused on military applications of AI and semiconductor export controls. At a time when the tech rivalry between Washington and Beijing is intensifying, suspected Chinese state-aligned cyber operatives have reportedly launched a targeted social-engineering campaign, posing as former US government officials and leading private-sector researchers to steal sensitive insights on American artificial intelligence (AI) policy.
On February 26, the attackers sent an email titled “Request for Feedback on Military Integration of Claude” to an AI analyst at a US think tank, attempting to harvest login credentials. The targets included analysts, legal counsel, and academic scholars across American universities, national security think tanks, and law firms, the findings claim (via CNN). Gaining access to the personal accounts and correspondence of these experts may hand Beijing an inside look at how American institutions are debating national security guardrails, trade curbs and regulatory frameworks around cutting-edge machine learning. The cyber operators deployed targeted deception campaigns across two distinct windows in February and July, adopting the identities of prominent figures in the artificial intelligence sector. These cybercriminals reportedly also impersonated a senior team member at AI startup Anthropic, which had recently sparred with the White House over deployment guardrails.
They posed as a former White House tech advisor, fabricating communications purporting to come from Lynne Parker, who previously served as a senior White House technology policy director under both Donald Trump and Joe Biden.