Technical corrections only. The changes between v3 and v4 reflect the bill's procedural progression from the House of Representatives to the Senate, including an updated attestation and referral information. There are no substantive policy changes to the legislative text.
Generative AI Terrorism Risk Assessment Act
- Sponsor
- Rep. Pfluger, August [R-TX-11]
- Committees
- Homeland Security Committee (primary)
- Last action
- Nov 20, 2025
Bottom line
This bill establishes a recurring federal assessment and information sharing framework to understand and counter the national security risks posed by generative AI when used by terrorist organizations.
What it actually does
This bill mandates the Secretary of Homeland Security, in consultation with the Director of National Intelligence, to conduct annual assessments for five years on terrorism threats to the United States posed by the use of generative artificial intelligence (AI). These assessments must analyze incidents where foreign terrorist organizations use generative AI to spread violent extremist messaging, radicalize individuals, or enhance their ability to develop or deploy chemical, biological, radiological, or nuclear weapons. The bill also requires enhanced information sharing among federal agencies and fusion centers regarding these threats.
Proponents argue
Supporters argue this bill is a crucial, proactive step to address an emerging national security threat, ensuring that federal agencies are systematically tracking and responding to the evolving capabilities of terrorists using advanced AI. They contend that by mandating regular assessments and fostering information sharing, the United States can better anticipate and mitigate risks to its people and infrastructure from AI-enabled terrorism.
Opponents contend
Critics might argue that existing intelligence frameworks are already sufficient to address evolving threats, and that mandating specific annual reports could create unnecessary bureaucratic burden or duplicate efforts. They might also contend that such prescriptive requirements could divert resources from more agile, real-time counter-terrorism operations or that the rapid pace of AI development could render fixed reporting mandates quickly outdated.
The bill is concise and easy to read, allowing for quick comprehension of its provisions.
Section 3(a)
Annual Assessments on Generative AI Terrorism Threats
This provision mandates the Secretary of Homeland Security, in consultation with the Director of National Intelligence, to submit annual assessments to Congress for five years. These assessments must analyze incidents during the preceding calendar year where foreign terrorist organizations or individuals used generative AI to spread violent extremist messaging, radicalize, recruit, or enhance their ability to develop or deploy chemical, biological, radiological, or nuclear (CBRN) weapons. It also requires recommendations on appropriate measures to counter these identified threats.
Supporters argue
Supporters argue this provision is essential for proactively identifying and understanding the evolving threat landscape, ensuring that policymakers have the necessary intelligence to develop effective counter-terrorism strategies against AI-enabled threats. They believe a systematic approach is vital given the rapid advancements in AI.
Critics contend
Critics might argue that such specific mandates could lead to redundant reporting if existing intelligence gathering and threat assessment processes already cover these areas, potentially diverting resources from more agile, real-time threat analysis. They may also question the ability of annual reports to keep pace with rapidly evolving technology.
Tradeoffs
This provision balances the need for specific, recurring threat assessments to inform policy against the potential for bureaucratic overhead and the challenge of keeping pace with rapidly evolving AI technology within a fixed reporting structure.
Section 3(b)
Enhanced Information Sharing on AI Terrorism Threats
This section requires the Secretary of Homeland Security to review and incorporate information relating to generative AI terrorism threats from State and major urban area fusion centers and the National Network of Fusion Centers. It also mandates the dissemination of relevant information back to these fusion centers. Furthermore, it requires the Office of the Director of National Intelligence, the Federal Bureau of Investigation, and other members of the intelligence community to share information regarding these terrorism threats with the Secretary of Homeland Security.
Requires the public version of the annual threat assessment to exclude information designated as 'For Official Use Only' or otherwise exempt from public disclosure under the Freedom of Information Act (FOIA) or other applicable law.
Section 3(a)(4)
Why it matters:This is standard legislative practice to protect sensitive national security information, intelligence sources, and methods, as well as privacy, while still allowing for some public transparency. It is not necessarily evasive but defines the limits of public access.
Case for: Proponents argue this is necessary to protect classified intelligence, sensitive operational details, and individual privacy, ensuring that the public release of information does not compromise national security or ongoing investigations.
Case against: Critics might argue that broad exemptions, even standard ones, can limit the public's ability to fully understand the nature and scope of AI terrorism threats and the government's response, potentially reducing accountability and hindering informed public discourse.
Estimated impact: Ensures that while some information is publicly available, the most sensitive details remain protected, impacting the scope of public discourse and oversight regarding AI terrorism threats.