How TA419 Used AI Collaboration Pitches to Target Policy Experts

Proofpoint says TA419 has tried to steal credentials from people working on AI policy since 2025. The campaign targeted specialists in areas such as AI regulation, export controls and national strategy, using emails that impersonated experts and proposed AI-related projects. The narrow focus points toward an interest in policy information, though the reports do not establish that the attempts led to successful account access. 17

Who was targeted, and when?

Proofpoint told Reuters it had observed TA419 attempting to steal passwords since 2025 from people at think tanks, defense contractors, universities and law firms in the United States and Japan. The targets included experts working on AI regulation, export controls and national AI strategy, as well as military applications of AI. 17

That makes the campaign distinct from indiscriminate phishing: its reported targets were people whose work could give them insight into AI policy and national strategy. Proofpoint characterized the targeting as limited and focused on a small number of people and organizations.

How the impersonation emails worked

The emails appeared to come from AI or policy experts, including former White House official Lynne Parker. They used proposed AI collaborations or initiatives as a lure, then directed recipients to websites intended to capture their passwords. Access to an account could in turn expose its email, but the reporting describes attempted theft rather than confirming that attackers obtained specific victims’ messages. 19

What happened to Alex Engler?

Alex Engler, a former White House official, told Reuters he received an email from someone posing as Parker and inviting him to join a new AI policy project. The message seemed suspicious to him; after checking with others in the field, he realized the sender was an impostor. The available reporting does not say that he entered credentials or that his account was compromised. 1

What evidence pointed to Chinese hackers?

Proofpoint attributed the activity to a Chinese group based on the malware used, the internet infrastructure involved and the targets it observed—indicators it assessed as consistent with Chinese intelligence priorities. This is Proofpoint’s investigative attribution, not evidence that every attempted intrusion succeeded. 1

What the campaign’s focus may suggest

The concentration on people working on AI regulation, export controls and national strategy suggests an interest in insight into policy decisions and plans, rather than simply access to AI systems. That is an inference from the campaign’s reported targeting and lures, not a confirmed account of the attackers’ goals or of what information, if any, they obtained. 7