Warner AI Bills Target Market Manipulation, Consumer Agents

Sen. Mark Warner (D-Va.) has introduced a legislative package that would expand federal oversight of artificial intelligence across financial markets, banking, consumer services, critical infrastructure and national security. In a speech on the Senate floor, he said Congress needs to establish regulatory guardrails before the technology becomes too deeply embedded in the economy to govern effectively.

Warner, who sits on the Finance, Banking and Budget committees, framed the package around four priorities: responsible infrastructure development, competition and safety, workforce disruption and national security. The package includes six new bills alongside previously introduced measures covering AI agents, financial fraud, data centers and frontier-model cybersecurity.

The financial-services implications are significant. Warner said he plans in coming weeks to reintroduce the bipartisan Financial Artificial Intelligence Risk Reduction Act. That bill would require financial regulators to address AI-generated content capable of disrupting financial markets.

“As AI-generated content becomes increasingly sophisticated, we also have to protect our financial markets and our democratic institutions from fraud, deception, and manipulation,” Warner said in his Senate floor speech.

The proposal reflects growing concern that realistic AI-generated text, images, audio or video could be used to manipulate markets. Deepfakes impersonating executives or government officials could spread false information about financial institutions, triggering rapid market reactions before fraudulent content can be authenticated.

Banking and payments could also be affected by Warner’s planned Artificial Intelligence Access, Gatekeeper Exchange, and Nondiscriminatory Transfer Act, or AI AGENT Act. The measure would establish rights and responsibilities for consumer-facing AI agents – systems that increasingly act autonomously when consumers shop, communicate and conduct financial transactions.

“If AI is going to become the interface through which Americans shop, bank, communicate, receive medical advice, and navigate the internet, then people need confidence that these systems are working for them – not exploiting them,” Warner told the Senate.

Under the proposal, AI agents accessing sensitive information such as email, eCommerce accounts and payment credentials would operate under fiduciary-like obligations to protect users’ interests. Trusted agents would receive a framework for accessing major online platforms while meeting privacy and cybersecurity standards. The National Institute of Standards and Technology would develop technical standards, while the Federal Trade Commission would maintain a registry of trusted agents.

That approach could have important implications for banks, payment providers and fintech platforms if autonomous agents increasingly initiate transactions or interact with financial accounts on consumers’ behalf. It would add new regulatory questions involving authentication, data access, liability, cybersecurity and responsibility for transactions initiated by AI.

Warner’s agenda also addresses the infrastructure supporting AI. His Data Center Tax Accountability and Disclosure Act would require large AI data centers to disclose information about electricity and water consumption, emissions, backup generation and other operational impacts. It would condition federal bonus depreciation on compliance with efficiency and sustainability standards. Revenue generated by limiting that tax benefit would help finance a National Workforce Transition Fund.

Warner warned that AI-driven disruption could hit white-collar occupations particularly hard, citing jobs in business, finance, marketing, software development and customer support as potentially vulnerable. The workforce fund would support retraining, portable training accounts, employer redeployment programs and tuition assistance.

National security forms another major pillar. The Secure AI Development Act would mandate secure pre-deployment testing of advanced AI models, improve government-industry reporting of AI cybersecurity vulnerabilities and establish voluntary incident reporting modeled on aviation safety systems. The National Security Agency would lead testing of frontier models, while federal vulnerability databases and processes would be updated to account for AI-specific risks.

Warner cast the package as an attempt to avoid repeating what he views as Congress’ failure to respond early enough to social media.

“We largely stood on the sidelines as social media reshaped our politics, our economy, and our children,” he said. “We cannot afford to make that same mistake again.”

The objective, Warner said, is “not slowing innovation but guiding it” – establishing federal rules before AI’s effects on markets, financial services, employment and national security become substantially harder to contain.