US House advances federal AI security center

Legislative progress

The U.S. House Science Committee advanced legislation that would create a federal artificial intelligence security center focused on evaluating AI systems used by federal agencies and, potentially, government contractors, according to Roll Call coverage cited by El Solitario.

The bill still has a long path ahead. It has cleared committee review, but it must still pass the full House, gain a matching version in the Senate, and finally receive the president’s signature before it can become law.

Gold Eagle is already underway

The legislative push is unfolding alongside Gold Eagle, the White House’s AI-based cybersecurity coordination system, which was introduced to detect and correct vulnerabilities in U.S. critical infrastructure through a clearinghouse model for information sharing between federal agencies and private companies.

According to Moncloa.com, Gold Eagle was authorized by Executive Order 14409, signed on June 2, 2026, and involves the Treasury Department, the Department of Homeland Security, CISA, and the Department of War as key players in critical infrastructure cybersecurity coordination.

Mitrade added that the program was publicly launched as an operating coordination system for AI and cybersecurity aimed at software vulnerabilities in critical infrastructure, with explicit coverage of financial systems. The same report described Gold Eagle as a system that classifies vulnerability reports submitted by both federal agencies and private companies, and coordinates prioritized remediation of those flaws.

Regulatory framework and third-party reach

Regulatory analyses from July 2026 place Gold Eagle and Executive Order 14409 within a broader voluntary AI-based cyber defense clearinghouse framework, with no comprehensive federal AI statute yet in place that would impose direct obligations on foreign vendors, according to Vorplabs.

Astraea Law said the U.S. regulatory architecture relies on executive orders and federal procurement memorandums, with substantive obligations concentrated in state laws. Under that setup, a future federal AI security center would operate first as an evaluation and coordination mechanism rather than as a sanctions-based regulator.

For Latin American subsidiaries and vendors that work with federal agencies or U.S. critical infrastructure, the legislative move and the launch of Gold Eagle point to an environment where risk reviews, vulnerability classification, and remediation coordination could become operational requirements before a comprehensive federal AI law exists.