AI policy

The amendments update the requirements for reporting, incident disclosure, and safety documentation for developers of large AI systems operating in New York State.

New York Governor Kathy Hochul signed NY Senate bill 8858 that amended the Responsible AI Safety and Education Act (RAISE Act), expanding disclosure and safety requirements for companies developing or deploying large-scale AI systems.

Under the revised law, “frontier” developers must now submit more detailed public disclosures about their models. Specifically – 

1. Detailed product documentation – This includes training methods, system capabilities, known limitations, and material updates to the product. 
2. Pre-deployment safety testing – Companies must test their models to ensure they’re safe enough for internal use at scale or for public release. Results must be reported.
3. Incident reporting – Developers must report “safety incidents” to the state within 72 hours. These include cases where an AI system causes or materially contributes to unauthorized system access, data exposure, or other harmful outcomes. Companies must provide details on what occurred, how it was detected, and what corrective actions occurred.
4. Internal safety and risk management documentation – Companies must document how they identify, evaluate, and mitigate risks throughout the AI system’s lifecycle. 
5. Testing documentation – The law requires companies to describe how they test systems for harmful or unintended behavior and what safeguards are in place to reduce those risks.
6. Show compliance – Retain records demonstrating compliance with the law’s requirements.

The amendments clarify that these obligations apply to developers of “frontier” AI systems, defined based on computational scale and advanced capabilities. Covered entities must ensure that disclosures remain current when material changes are made to the system.

The amendments go into effect January 1, 2027.

The legislation seeks to regulate “automated employment-related decision systems,” defined as AI tools used in hiring, evaluation, and termination decisions.

The State of Connecticut’s General Assembly’s Labor and Public Employees Committee introduced Senate Bill 435 to establish requirements for the use of AI in employment decisions.

The legislation would prevent the use of AI systems in ways that result in unlawful employment discrimination and require oversight of their use. It applies specifically to “automated employment-related decision systems,” which the bill defines as computational processes, including AI and machine learning, that make or materially assist in making decisions about employment hiring, evaluation, and termination.

The bill would require employers to conduct impact assessments before using these systems. These assessments must evaluate whether the system creates a disparate impact based on protected characteristics, including race, color, religion, sex, sexual orientation, gender identity, national origin, age, or disability. Employers must also document how the system works, what data it uses, and how risks are identified and addressed.

It also mandates ongoing testing and auditing of these systems after deployment. Employers would need to self-monitor outcomes and take corrective action if they identify discriminatory patterns.

The legislation also requires transparency, requiring employers to notify employees and job applicants when using an automated system in connection with an employment decision. Employers would also provide candidates and employees with information about how the system contributed to decisions that affect them.

The bill outlaws automated systems that lead to unlawful discrimination, regardless of whether the discrimination was intentional.

For AI companies, the bill indirectly forces compliance requirements. Companies that sell AI hiring or workforce management tools would need to ensure their systems can be tested for bias, audited for outcomes, and explained to end users, since employers would be required to meet those standards in practice.

The bill has been referred to committee and remains under consideration in the Connecticut General Assembly.

The bi-partisan legislation would create federal liability and enforcement measures targeting AI-enabled impersonation and related fraud.

House members Vern Buchanan (R-Fla.) and Darren Soto (D-Fla.) introduced H.R. 7786, titled the AI Fraud Accountability Act, to establish federal protections against digital impersonation fraud, including cases involving AI-generated content.

Under the legislation, individuals who create or use digital replicas or impersonations of another person without consent, and with intent to defraud or cause harm, could face civil liability. Under the bill, individuals may file civil lawsuits against violators, seeking to take down any harmful material and sue for monetary damages.

The legislation also sets rules for online platforms and services that host or distribute this kind of material. To limit their liability, these companies would need to act once notified of an impersonation. That includes removing the content and taking reasonable steps to prevent it from being reposted.

The bill assigns enforcement authority to federal agencies by treating these impersonation schemes as violations subject to existing federal oversight. In practice, this allows Federal agencies to pursue cases where impersonation is used in connection with deceptive or fraudulent conduct, using their existing powers to investigate and enforce fraud cases involving AI-generated replicas.

The legislation also defines key terms, such as what constitutes a “digital replica,” to ensure that courts apply a consistent standard when evaluating cases.

The requirements would apply broadly to individuals who create the content, as well as to platforms and service providers that distribute or enable it.

H.R. 7786 has been introduced and referred to multiple House committees, including the Committees on Energy and Commerce, the Judiciary, Science, Space, and Technology, and Foreign Affairs, for further consideration.

No floor votes or additional legislative actions have been recorded at this stage.

The White House released a seven-point recommendation calling for Congress to establish a national policy framework for AI.

The document outlines seven key categories where the administration wants federal lawmakers to focus on in establishing federal legislation governing AI.

These are: 

(1) protecting children and empowering parents.

(2) safeguarding communities.

(3) respecting intellectual property rights and supporting creators.

(4) preventing censorship and protecting free speech.

(5) enabling innovation.

(6) developing an AI-ready workforce.

(7) setting rules for federal and state authority over AI.

The federal standards call for preempting some state AI laws while preserving state authority in areas such as child protection, fraud, consumer protection, zoning, and a state’s own use of AI.

The document states the federal government “must establish a federal AI policy framework to protect American rights, support innovation, and prevent a fragmented patchwork of state regulations that would hinder our national competitiveness, while respecting federalism and State rights.”

The recommendations also call for federal efforts to stimulate innovation by creating federally run sandboxes, granting access to federal datasets for AI model training, and ensuring workforce training via existing education and support programs. They further say Congress should not create a new federal AI rulemaking body and should instead rely on existing regulators with subject-matter expertise and industry-led standards.

Draft regulation sets out how the European Commission would assess general-purpose AI models and conduct enforcement proceedings under the AI Act.

The European Commission published a draft regulation outlining how it would conduct evaluations and enforcement proceedings for general-purpose AI models under Regulation (EU) 2024/1689, the EU Artificial Intelligence Act.

The proposal, available via the Commission’s “Have Your Say” portal, would establish detailed procedures for evaluating AI models, appointing independent experts, and enforcing compliance, including the potential imposition of fines.

Under the draft, the Commission would be empowered to require providers of general-purpose AI models to grant access for evaluation purposes, including through application programming interfaces, source code, model weights, and infrastructure. Providers would be required to provide such access “in a timely and effective manner” to support the Commission’s assessment objectives.

The regulation would also formalize the role of independent experts. The Commission would select experts through open procedures and assess their independence based on factors such as prior relationships with AI providers. Experts would be required to maintain confidentiality and adhere to information security standards.

In enforcement matters, the draft sets out procedures for opening and closing proceedings against providers of general-purpose AI models. The Commission would be able to initiate proceedings where it suspects violations of the AI Act and, in urgent cases, impose interim measures, including restricting a model’s availability on the market.

The proposal includes procedural safeguards for companies under investigation. Providers would have the right to respond to preliminary findings in writing, with a minimum response period of 14 days. The Commission would also be required to grant access to relevant case files, subject to protections for confidential business information.

The draft introduces limitation periods, including a five-year window for the Commission to impose fines and a separate five-year period for enforcement of penalties.

The regulation has not yet been adopted and reflects preliminary views of Commission services. It is open for feedback through the European Commission’s consultation process until April 9th, 2026.

The proposed framework would establish limits on military AI use and domestic surveillance amid ongoing disputes over federal AI oversight

Senator Adam Schiff (D-Calif.) is drafting legislation to establish guardrails on the use of AI in autonomous weapons and domestic surveillance, according to a March 13, 2026, press release from his office.

The proposal, in its early drafting stages and not yet formally introduced, would seek to limit how federal agencies deploy AI in national security and law enforcement contexts. It would apply broadly to federal departments, including defense and intelligence agencies, and is expected to address both battlefield applications and the use of AI for monitoring individuals within the United States.

In a statement accompanying the announcement, Schiff said, “Commonsense safeguards are critical to protect the privacy, values, and national security interests of the American people, and after consulting with industry and experts, I’ll be introducing legislation in the coming weeks to codify vital protections.”

The draft effort follows recent tensions between the Department of War and Anthropic over restrictions on the company’s technology in government systems. The fallout came to a head in early March when the Department of War designated Anthropic as a supply chain risk, followed by a lawsuit filed by Anthropic against the Department of War.

Schiff’s office said the legislative effort is being developed in consultation with other lawmakers, though no co-sponsors have been formally announced. 

The proposal remains in draft form and has not been introduced in either chamber of Congress. No timeline for formal introduction or committee consideration has been provided.

The proposed legislation would establish a federal commission to study how artificial intelligence affects American workers and the economy.

Senator Mark Warner (D-Va.) and Senator Todd Young (R-Ind.) introduced the Economy of the Future Commission Act, a bill that would create a federal commission to examine the impact of artificial intelligence (AI) and other emerging technologies on the U.S. workforce and economic competitiveness.

The proposed commission would be tasked with studying how AI is expected to affect employment, wages, productivity, and job displacement across industries. It would also review workforce training needs, education systems, and federal policies related to technological change. The commission would be required to produce a report to Congress with findings and recommendations within a specified timeframe.

Under the bill, the commission would include members appointed by congressional leadership and the executive branch, drawing from government, industry, labor, and academic backgrounds. It would hold hearings, collect data, and consult with relevant stakeholders to inform its analysis and recommendations.

In a statement accompanying the bill, Senator Mark Warner (D-Va.) said, “We need clear-eyed analysis and practical, bipartisan solutions to help workers gain new skills, support people whose jobs are disrupted, and make sure the United States leads the world in the industries of the future.” Senator Todd Young (R-Ind.) added, “America’s workforce must be equipped to lead the transformation of the economy happening due to AI. This commission would help keep America ahead of our global competitors and keep America prosperous and innovative.”

The legislation would apply broadly to federal policymakers and agencies by informing potential future legislation or regulatory actions related to AI and workforce development. It does not impose new regulatory requirements but is intended to guide congressional decision-making.

The bill has been introduced in the Senate and referred to the appropriate committee for consideration.