The White House Now Uses AI to Hunt Software Vulnerabilities

The Trump administration has stood up Gold Eagle, a federal clearinghouse that uses frontier AI models to find, prioritize, and coordinate patching of software vulnerabilities across government and private industry.

The program, led by the Treasury Department with support from CISA, the Pentagon, and open-source software partners, was created under a June 2, 2026 executive order and is already operational.

A senior White House official confirmed that closed-source AI models are participating in Gold Eagle’s vulnerability-hunting work, including Anthropic’s Claude Mythos, a model Anthropic specifically designed for advanced cybersecurity research, as Fox News reported.

Anthropic has said Mythos-class models can both find vulnerabilities and develop working exploits for them — capabilities the company acknowledges could make attacks easier if the models fall into the wrong hands.

How the system actually works

At the core of Gold Eagle is VINCE — the Vulnerability Information and Coordination Environment — operated in partnership with Carnegie Mellon University’s Software Engineering Institute.

VINCE lets anyone report a vulnerability for triage and coordinated remediation, aiming to compress the traditional disclosure timeline by having AI models handle the discovery, verification, and prioritization work that previously required extensive manual security research.

The tension nobody has fully resolved

Gold Eagle is explicitly a response to the same dynamic playing out elsewhere this month: AI models are now finding real zero-day vulnerabilities faster than human researchers can, for better or worse.

The same capability that lets Gold Eagle patch flaws at unprecedented speed is functionally identical to what a malicious actor could use to find and weaponize those flaws.

National Cyber Director Sean Cairncross has called on Congress to reauthorize the CISA 2015 Act, which expires in September and on which officials say the entire program legally depends — a deadline that could interrupt Gold Eagle’s operations if lawmakers don’t act.

The program is notably narrower in one respect than the industry-led Open Secure AI Alliance that Nvidia launched the same week: Gold Eagle is a government-operated clearinghouse that participating labs feed into, while the Alliance is a peer coalition building shared open tools outside of any single government process.

Officials have not disclosed a full list of which AI companies beyond Anthropic are contributing models or capacity to Gold Eagle, leaving open how broad its industry participation actually is.

What to watch next

• Whether Congress reauthorizes the CISA 2015 Act before its September expiration.
• Which other AI labs beyond Anthropic formally join Gold Eagle’s vulnerability-hunting effort.
• Whether Gold Eagle’s speed advantage measurably reduces the window between vulnerability discovery and patching across participating industries.

Helpful Sources

• Trump launches Gold Eagle to hunt cyber flaws with AI — Fox News
• Trump administration unveils AI-supported clearinghouse for cyber vulnerabilities — The Record
• US launches vulnerability clearinghouse amid AI-fueled surge in flaws — Cybersecurity Dive

Disclaimer: This content is meant to inform and should not be considered financial advice. The views expressed in this article may include the author’s personal opinions and do not represent Times Tabloid’s opinion. Readers are advised to conduct thorough research before making any investment decisions. Any action taken by the reader is strictly at their own risk. Times Tabloid is not responsible for any financial losses.

Follow us on X, Facebook, Telegram, and  Google News

The post The White House Now Uses AI to Hunt Software Vulnerabilities appeared first on Times Tabloid.