Sen. Mark Warner, Vice Chairman of the Senate Intelligence Committee (with Sens. Hawley, Rounds, Young, and others) · 2026
Sen. Mark Warner — Comprehensive AI Legislative Agenda
Warner
Rolled out July 21, 2026, Warner's agenda is the most legislatively dense package any single member of Congress has put behind AI — seven bills organized around four stated priorities: building AI infrastructure responsibly, promoting competition and safety, preparing workers for economic disruption, and strengthening America's national security advantage. The framing is explicitly anti-tradeoff: "Artificial intelligence will reshape nearly every aspect of our economy and society. The only question is whether Congress is going to help shape that future or spend the next decade scrambling to catch up." On infrastructure, the Data Center Tax Accountability and Disclosure Act forces large AI data centers to publish energy, water, emissions, and backup generation data, and conditions federal bonus depreciation on meeting efficiency standards — with the revenue from limiting that depreciation dedicated to a National Workforce Transition Fund. On competition and safety, the AI AGENT Act builds a rights-and-duties regime for consumer-facing AI agents accessing major platforms, and the SAFE AI Act bans federal procurement of models that generate CSAM or non-consensual intimate imagery while creating private remedies for survivors. On national security, the Secure AI Development Act mandates secure testing environments for advanced models before deployment and imports aviation-style voluntary incident reporting into AI. The agenda builds on Warner's earlier workforce measures — the AI-Related Job Impacts Clarity Act with Sen. Hawley, the Economy of the Future Commission with Sen. Rounds, and the Investing in American Workers Act — converting what had been a data-and-commission strategy into a full regulatory program.
Key Provisions
Data Center Tax Accountability and Disclosure Act: mandatory public disclosure of energy consumption, water usage, emissions, and backup generation by large AI data centers, with federal bonus depreciation benefits conditioned on meeting efficiency and sustainability standards (backed by the Chesapeake Bay Foundation, Sierra Club, and Virginia League of Conservation Voters)
National Workforce Transition Fund: dedicates the revenue from limiting data center bonus depreciation to worker retraining, individual training accounts, employer retention grants, tuition assistance, workforce data modernization, and pilot programs — an explicit pay-for linking AI infrastructure subsidies to AI displacement
AI AGENT Act (Artificial Intelligence Access, Gatekeeper Exchange, and Nondiscriminatory Transfer Act): establishes rights and responsibilities for consumer-facing AI agents, permitting trusted agents to access major platforms while requiring them to act in the user's interest, secure personal data, and meet privacy and cybersecurity standards; directs NIST to write the technical standards and creates an FTC registry
SAFE AI Act (Safeguarding Against Fabricated Exploitation Through Artificial Intelligence Act): NIST-led testing of AI models, a ban on federal procurement of systems that generate child sexual abuse material or non-consensual intimate images, prohibition of circumvention software, and private legal remedies for survivors
Secure AI Development Act: mandatory secure testing environments for advanced models before deployment, modernized federal cybersecurity vulnerability identification, structured government–developer information sharing, voluntary safety incident reporting modeled on aviation protocols, and hardened protections against foreign adversaries targeting AI technologies and supply chains
Financial Artificial Intelligence Risk Reduction Act (bipartisan, to be reintroduced): requires financial regulators to address the risk that AI-generated content disrupts markets
PHD Talent Act: expands undergraduate-to-doctoral pathways in AI, cybersecurity, semiconductors, and quantum computing (endorsed by Virginia Tech, Purdue, UVA, and the Southeastern Universities Research Association)
AI-Related Job Impacts Clarity Act (with Sen. Hawley): quarterly disclosures from publicly traded companies and federal agencies to the Bureau of Labor Statistics on AI-driven layoffs, hires, unfilled positions, and retraining, each tagged with NAICS codes, with rulemaking authority to extend reporting to large private companies
Economy of the Future Commission Act (S.4046, with Sen. Rounds): a 10-member bipartisan legislative commission with subpoena power, required to deliver 5- and 10-year AI employment projections by 6-digit NAICS code within 7 months and consensus legislative recommendations within 13 months
Investing in American Workers Act: employer tax credits for worker training
Bipartisan letter (Warner, Hawley, Banks, Hassan, Kelly, Kaine, Hickenlooper, Young, Rounds) directing BLS and Census to add AI questions to the CPS, JOLTS, and the National Longitudinal Survey
Regulatory Philosophy
Regulate the seams, not the models. Warner's agenda conspicuously declines the two positions that dominate the federal debate — it neither imposes a licensing or liability regime on frontier developers nor endorses preemption of the states. Instead it attaches obligations at the points where AI touches something Congress already regulates: the tax code (bonus depreciation for data centers), federal procurement (the CSAM and NCII ban), platform access (the agent registry), financial market integrity, immigration and education pipelines, and classified national security infrastructure. The pay-for structure is the tell — the same provision that limits a tax subsidy for AI infrastructure funds the retraining of the workers that infrastructure displaces, making the agenda internally financed rather than dependent on new revenue. As Vice Chairman of the Intelligence Committee, Warner also treats AI security as a supply chain and counterintelligence problem rather than an alignment problem, which is why the Secure AI Development Act reaches for aviation's incident-reporting model instead of a safety regulator.
Data center bonus depreciation, conditioned on efficiency standards
Why it lands there
The clearest hyperscaler-incidence proposal in the landscape, and the only one that names its base precisely. Conditioning bonus depreciation on efficiency standards and dedicating the recovered revenue to worker transition assesses the layer that owns the buildings and the machines. The AI AGENT Act reaches the app and platform layer through gatekeeper access rules, and secure testing environments reach the frontier, but the money comes from capacity.
+Seven bills spanning tax, procurement, competition, workforce, education, financial regulation, and national security make this the broadest legislative agenda from any single member — it addresses infrastructure, environment, and security gaps that most AI proposals in this landscape leave entirely blank
+Conditioning bonus depreciation on efficiency standards and dedicating the recovered revenue to worker transition is the rare AI worker proposal that identifies its own funding source, avoiding the unfunded-aspiration problem that afflicts the Kelly Horizon Fund and most commission-based approaches
+Mandatory data center disclosure of energy, water, and emissions is the most concrete response yet to the environmental footprint of AI infrastructure, an area this landscape otherwise treats only glancingly
+The AI AGENT Act anticipates a market structure problem before it hardens — whether platforms can exclude third-party agents — rather than litigating it a decade later, and pairs it with NIST standards and an FTC registry instead of a new regulator
+The procurement lever in the SAFE AI Act is enforceable without new regulatory machinery: the federal government simply declines to buy, which sidesteps the constitutional and administrative fights that a content mandate would trigger
+Warner's Intelligence Committee position gives the Secure AI Development Act credibility on classified testing environments and adversary supply chain threats that a non-committee sponsor could not claim, and the aviation incident-reporting model is a proven design for surfacing near-misses without punitive exposure
Weaknesses
From the perspective of political opposition
−Seven separate bills across at least five committees of jurisdiction is a strategy for producing press releases, not laws — the agenda has no vehicle, no floor path, and no named Republican co-sponsors on the new measures, and its most substantive pieces will die in committee separately
−Regulating at the seams means never regulating the thing itself: nothing in the agenda constrains what a frontier model may do once it is deployed commercially, so the entire question of catastrophic capability is handed to a voluntary reporting scheme
−Voluntary safety incident reporting modeled on aviation misreads the analogy — aviation's system works because the NTSB has subpoena power, mandatory accident reporting, and grounding authority, none of which this bill provides
−Conditioning tax benefits on efficiency standards invites exactly the compliance theater the data center industry has already perfected with renewable energy credits and water-use accounting, letting firms claim standards while shifting environmental costs onto host communities and ratepayers
−The agenda is silent on federal preemption, the deepest fault line in AI policy, which is a choice that allows Warner to avoid alienating either state attorneys general or the industry but leaves the central jurisdictional question unresolved
−Warner's earlier workforce package remains what it always was — a commission that will spend 13 months studying displacement while it happens, and quarterly corporate self-reports that companies will game by classifying AI-driven layoffs as ordinary restructuring
−For a package built on disclosure and accountability, exempting the Economy of the Future Commission from FACA and FOIA is a conspicuous carve-out from the transparency the agenda demands of everyone else