What You Need to Know
- Key takeaway #1The new Democratic majorities in Congress will likely press for legislation on data centers, frontier AI model safety, cybersecurity, workers’ welfare, civil rights, and national security.
- Key takeaway #2Bipartisan interest in several areas could create openings for legislation, but divided government makes the scope and timing of any laws uncertain.
- Key takeaway #3Congress may move in the lame-duck session to restrict AI chip exports.
Concerns about artificial intelligence played a seminal role in the midterm congressional elections. One exit poll found that 56% of voters said that AI and data centers were a “crucial or major factor” in their vote for Congress, and voters passed over 40 ballot measures to restrict data centers.
While it remains to be seen what proposals can become law with the Congress controlled by one party and the White House by another, incumbent and incoming Democratic and Republican lawmakers are floating a raft of initiatives to scrutinize data center construction, impose AI safety standards on frontier models, promote workers’ welfare, protect civil rights, and protect national security. These legislative efforts will likely complement lawmaking at the state level and state and federal oversight and investigations of AI and technology.
Top Policy Priorities
Several key issues have emerged that will likely define the AI legislative agenda during the 120th Congress.
Data Centers
Given the negative public perception of data centers, we expect the new Congress to introduce bills to mitigate the environmental and energy impacts of data center construction.
For example, Sen. Mark Warner (D-VA) introduced in July 2026 the Data Center Tax Accountability and Disclosure Act to require data centers to disclose information about their emissions and energy and water consumption and to condition tax benefits on meeting efficiency and sustainability standards.
Concerns about data centers’ resource use resonate on both sides of the aisle. In September 2026, Sen. Shelley Moore Capito (R-WV), among others, introduced the Bipartisan American Affordability and Jobs Act (BAAJA) to modify the federal permitting processes and, among other things, direct the Federal Energy Regulatory Commission (FERC) to establish ratepayer protections related to data center energy consumption and transmission costs. On November 9, the Senate voted 73 to 25 to advance BAAJA, which now will undergo an amendment process before the Senate votes on final passage.
In the House, Rep. Greg Landsman (D-OH-01) introduced the No Harm Data Centers Act in April 2026, which would force data centers to pay the full cost of their energy and infrastructure demands.
Incoming members of Congress promise to inject new energy into this issue; Rep. -elect William Lawrence (D-MI-07) centered his candidacy on opposition to local data center development and called for a pause on AI development.
Frontier Model Safety
Democrats and Republicans both support bills to require frontier AI developers to establish a compulsory risk-based framework for governing the development and deployment of the most advanced artificial intelligence models.
For example, Rep. Jay Obernolte (R-CA-23) introduced in June 2026 the FRONTIER Act, which now has 11 bipartisan co-sponsors, to require AI developers to take specific steps to ensure model safety, including requiring the reporting of AI incidents, independent auditing, and ongoing safety assessments. The bill would also give the U.S. Secretary of Commerce authority to issue an “emergency order” to suspend or restrict a frontier developer’s development, deployment, or internal use of a frontier model if the Secretary finds that such activity “presents an imminent catastrophic risk.”
Also, in September 2026, Sen. Bernie Sanders (I-VT) and Rep. Greg Casar (D-TX-35) introduced in both houses the Ban Artificial Superintelligence Act, which would bar the release of AI superintelligence and pause advanced AI development until a federal agency is established to safeguard the public from AI risks. The bill defines superintelligence as the ability to accelerate AI research and development, the capacity to access secured digital or physical systems without authorization, and the ability to continue operating even if a human tries to shut it down, among other criteria.
Sen.-elect Abdul El-Sayed (D-MI) built off the Sanders-Casar proposal, offering his own framework to get AI Under Control. This framework also calls to ban AI superintelligence as well as recursive self-improvement — where AI systems themselves design more capable AI systems — and proposes an independent AI safety agency and mandatory breach reporting, among other measures.
News reports indicate that Rep. Ro Khanna (D-CA-17), who co-sponsored the Ban Artificial Superintelligence Act, plans to introduce the Human Control Over AI Act to prohibt models capable of recursive self-improvement until a host of federal guardrails are in place. The bill would criminalize “crimes against humanity,” meaning the deployment of AI models that result in the destruction of civilian populations, and impose criminal penalties on AI company employees who deploy unauthorized systems or disable technical safeguards.
Rep. Ted Lieu (D-CA-36) plans to push for the AI Kill Switch Act, which would require covered entities to “[m]aintain a technical capability” to shut down their AI systems. The bill would also give the U.S. Secretary of Homeland Security authority to order a shutdown of AI systems when they cause certain catastrophic incidents.
Sen. Mark Warner (D-VA), potentially the incoming Chairman of the Senate Select Committee on Intelligence, released in July 2026 a sweeping Framework for America’s AI Future. Sen. Warner’s framework proposes the SAFE AI Act to create a process led by the National Institute of Standards and Technology (NIST) to test and verify AI models and prohibit software meant to bypass AI safety controls. He also proposes the AI AGENT Act, which would allow verified AI agents to access major online platforms while acting in the best interest of consumers and protecting sensitive personal information.
Rep.-elect Scott Wiener (D-CA-11), who as a state legislator authored several California laws on AI, also laid out a broad AI policy agenda. He proposes reviving the U.S. Digital Service to advise federal agencies on the benefits and risks of AI and efforts to guard against AI catastrophic risk through federal regulation, transparency and safety requirements, independently verified safety standards, fines for violating those standards, and AI whistleblower protections.
Cybersecurity
Especially in light of recent AI agent-enabled cyber breaches, we expect the incoming Congress to focus on cybersecurity.
Some bills, such as Sens. Josh Hawley (R-MO) and Chris Murphy’s (D-CT) AI Agent Accountability Act and Rep. Lori Trahan’s (D-MA-03) CLAIM Act, would create causes of action to hold AI agent operators and developers liable for cybersecurity hacking incidents.
Also, the Stop Rogue AI Act, proposed by Rep. Josh Gottheimer (D-NJ-05), a co-chair of the House Commission on Artificial Intelligence and the Innovation Economy, would direct NIST to develop and maintain standards for the secure use of AI agents by an organization, including multiple levels of security measures, known as “defense in depth.”
The FRONTIER Act would require the reporting of cybersecurity incidents, a proposal echoed in an AI policy framework written by Sen. Maria Cantwell (D-WA). Sen. Cantwell’s framework requires that developers promptly report material AI safety and security incidents, including the loss of control over autonomous agents, dangerous cyber capabilities, failures of critical safeguards, and unsafe recursive self-improvement. The AI Flaw Reporting and Security Enhancement Act, sponsored by Rep. Deborah Ross (D-NC-02), would also facilitate the voluntary reporting and tracking of AI flaws and vulnerabilities.
Additional proposals urge public-private partnerships to deploy defensive AI to detect malicious activity and patch cyber vulnerabilities; the protection of critical infrastructure, with particular focus on equipping local governments and private sector operators with the training and resources to safeguard essential services; as well as mitigating cyber threats from emerging technology like quantum computing and national security adversaries like China.
Workers’ Welfare
Sen. Mark Warner proposed in July the National Workforce Transition Fund, which would direct revenue generated by the depreciation of AI data center infrastructure to a fund supporting workforce retraining, employer retention grants, tuition assistance, workforce data modernization, and pilot programs to navigate technological disruption. Sen. Warner also introduced the PHD Talent Act, which would enhance the supply of domestic STEM talent through the creation of accelerated pathways to doctorate degrees in critical technology fields.
In concert with Sen. Warner, Sen. Mike Rounds (R-SD) introduced the Economy of the Future Commission Act, aimed at identifying Congressional measures to support workers as the job market evolves due to AI. These include expected employment changes, resources for understanding the technology, reskilling, as well as taxation and unemployment insurance policy.
Sen. Mark Kelly’s (D-AZ) Make AI Work for Americans Act would establish a federal trust fund to support American workers to be funded by leading AI companies via taxes on excess profits, digital ad services, and AI usage.
New members are also seeking solutions to AI-driven labor displacement. Rep.-elect Manny Rutinel (D-CO-08) campaigned on requiring notice before AI-driven layoffs at large employers, and Rep.-elect Wiener highlighted the expansion of unemployment benefits to prepare for AI job loss as a key tenet of his platform.
Civil Rights
Shortly after the midterms, Rep. Ayanna Pressley (D-MA-07) urged Democrats to prioritize bills such as the AI Civil Rights Act, which would prohibit developers or deployers from offering covered algorithms that cause disparate impact or discrimination preventing the equal enjoyment of goods, services, or opportunities related to certain consequential domains on the basis of a protected characteristic.
The Congressional Black Caucus (CBC) has also vowed to ensure that civil rights protections are part of any AI bill that becomes law.
National Security
National security and competition with China remain high on the list of AI policy priorities. As part of his framework, Sen. Mark Warner’s Secure AI Development Act would require a secure testing environment for the most advanced AI models before deployment and enhance protections against the theft or exploitation of U.S. AI technologies and supply chains by foreign adversaries.
Rep. Khanna’s Human Control Over AI Act would regulate advanced chips used in frontier AI through mechanisms that monitor and control their use. The bill also calls for enforceable agreements and targeted export controls to deter China (and others) from developing dangerous AI systems.
In response to concerns about insufficiently strict export restrictions of advanced AI chips to China, lawmakers during the lame-duck session may include in the must-pass National Defense Authorization Act a trio of bills: the AI Overwatch Act, which would require the Under Secretary of Commerce for Industry and Security to require a license for the export and transfer of certain integrated circuits; the Chip Security Act, which would require the Commerce Secretary to issue standards with respect to chip security mechanisms such as location verification; and the MATCH Act, which would place export restrictions on certain semiconductor manufacturing equipment and ensure that allied supplier countries are aligned with such restrictions. Oversight of the national security implications of AI remains a bipartisan concern.
Implications
The incoming Congress is primed to engage in policymaking on a range of AI issues. Even if these bills do not become law, this heightened congressional activity could increase pressure on companies to demonstrate the safety, security, and fairness of their AI systems and defend the economic and environmental impacts of their investments. These efforts, which will be joined by state policymaking and congressional investigations, will further complicate the variegated state, federal, and international compliance environment for AI development and deployment.

For further information, please contact:
Matthew F. Ferraro, Partner, Crowell & Moring
mferraro@crowell.com




