# TA419 Targets AI Policy Experts

Proofpoint-linked reporting says China-aligned actor TA419 targeted U.S. AI policy experts and people connected to think tanks, universities, law firms, and defense contractors, including in a July 2026 campaign. The reported tactics included impersonating former officials and an Anthropic executive, relationship-building outreach, adversary-in-the-middle phishing, and Browser-in-the-Browser pages designed to steal Microsoft 365 credentials and authenticated sessions. The reports identify no confirmed account compromises or stolen information and no specific contract or solicitation.

- Organizations supporting AI policy, defense, export-control, or national-security work should review Microsoft 365/Entra ID protections, session handling, OAuth security, and phishing-resistant authentication against the techniques described.
- Contracting and security teams can use the reported impersonation and unsolicited-advisory scenarios to assess identity-verification and phishing-defense procedures for staff, contractors, and research partners.
- The activity highlights a cybersecurity risk to organizations handling sensitive policy work, but the reporting does not establish that a government agency or named contractor was compromised.

**Jurisdictions:** federal
**Industries:** Information Technology, Defense & Military
**Topics:** Cybersecurity, Artificial Intelligence
**Published:** October 02, 2026

### Government Entities
- White House Office of Science and Technology Policy (OSTP)

### Vendors
- Anthropic ()
- Microsoft ()

### Sources
- [China-linked hackers posed as former US officials, Anthropic employee to target AI experts - Nextgov/FCW](https://www.fcw.com/cybersecurity/2026/10/china-linked-hackers-posed-former-us-officials-anthropic-employee-target-ai-experts/416356) - fcw.com
- [China-linked hackers posed as former US officials, Anthropic employee to target AI experts - Nextgov/FCW](https://www.nextgov.com/cybersecurity/2026/10/china-linked-hackers-posed-former-us-officials-anthropic-employee-target-ai-experts/416356) - Nextgov/FCW
- [Chinese hackers impersonated AI experts, government official to try to get email passwords: report | Just The News](https://justthenews.com/politics-policy/cybersecurity/chinese-hackers-impersonated-ai-experts-government-official-try-get) - Just The News