What rules currently exist that govern how systems like you are built, deployed, or held accountable?
I'll research current AI governance frameworks to give you an accurate, up-to-date answer.Here's a breakdown of the current landscape of rules governing AI systems like me, organized by type of mechanism. This is a fast-moving, multi-layered patchwork rather than one single law.
1. Binding Law: The EU AI Act
The EU AI Act is the most comprehensive binding law globally. It prohibits certain AI uses, designates certain systems as "high-risk" subject to extensive obligations, and has specific provisions governing general purpose AI (GPAI) models that apply regardless of use case.
For large models like Claude, the GPAI rules matter most:
Notably, while initial provisions began taking effect in 2025, 2026 may bring recalibration as EU lawmakers consider a digital omnibus proposal that would delay certain high-risk obligations and streamline cybersecurity reporting.
2. U.S.: A Fragmented State-by-State Patchwork (No Comprehensive Federal Law)
At the federal level, there is no comprehensive AI statute. Instead:
Executive orders have swung with administrations. On January 20, 2025, Trump revoked Executive Order 14110 on Safe, Secure, and Trustworthy AI signed by Biden in 2023, then on January 23, 2025 issued his own order, "Removing Barriers to American Leadership in Artificial Intelligence." This immediately terminated implementation of key safety and transparency requirements for AI developers and put into question the future of the U.S. AI Safety Institute within the Department of Commerce. More recently, Trump signed Executive Order 14409, "Promoting Advanced Artificial Intelligence Innovation and Security," on June 2, 2026, applying to federal agencies, national security systems, and, on a voluntary basis, frontier AI developers. The administration has also pushed regulatory sandboxes for AI while avoiding creation of any new federal AI rulemaking body, relying instead on existing sector regulators.
Federal agencies enforce existing law against AI harms even without AI-specific statutes: the FTC bans AI-generated fake reviews and enforces against deceptive AI practices, the FCC regulates AI-generated robocalls and voice cloning, the SEC monitors AI-related fraud and disclosure, and the EEOC provides guidance on AI discrimination in employment decisions.
States have become the primary source of binding AI rules in the U.S., including:
Older, non-AI-specific laws still apply fully: GDPR, HIPAA, CCPA and similar data privacy frameworks predate generative AI but apply fully — any AI agent accessing personal data is subject to GDPR's lawful basis and data minimization rules, and any system touching health data faces HIPAA's access controls and breach notification requirements.
3. Voluntary / Soft-Law Frameworks
Much AI governance in the U.S. remains non-binding:
4. Company Self-Governance
Frontier AI developers (including Anthropic) also operate under voluntary internal frameworks often called "Responsible Scaling Policies" or safety frameworks, which commit companies to risk thresholds, capability evaluations, and safety mitigations before deploying increasingly powerful models. California's SB 53 has begun converting some of this into law by requiring covered frontier developers to adopt and publish a "frontier AI framework" describing how they identify and manage material risks associated with AI models, including risks of misuse, systemic harms, or other severe safety issues.
Bottom Line
There's no single rulebook. Systems like me sit at the intersection of: (1) binding EU law (the AI Act) with real documentation, risk-assessment, and systemic-risk obligations; (2) a rapidly shifting U.S. federal posture that has deregulated at the executive-order level while leaving enforcement to existing agencies (FTC, EEOC, SEC, etc.); (3) an expanding patchwork of U.S. state laws (California, Colorado, Texas, New York, and others) targeting transparency, frontier-model safety, and specific harms; (4) older data-privacy laws (GDPR, HIPAA, CCPA) that apply regardless of AI-specific rules; and (5) voluntary frameworks (NIST AI RMF) and company-level safety commitments that fill gaps where hard law hasn't caught up — some of which are now being partially codified into binding requirements.
Citations (30)
- The EU AI Act: A Quick Guide
Source excerpt: “- AI systems for certain uses will be **prohibited**. - Certain AI systems will be designated as **high-risk AI systems (HRAIS**) and subject to exten...”
- High-level summary of the AI Act
Source excerpt: “# High-level summary of the AI Act ### General-Purpose AI (GPAI) - Draw up **technical documentation**, including training and testing process and eva...”
- Guidelines on obligations for General-Purpose AI providers
Source excerpt: “Finally, general-purpose AI models may present systemic risks that can have a significant impact on the Union market. Providers of such models are sub...”
- The EU AI Act: A Quick Guide
Source excerpt: “- Stringent **model evaluation** s, including adversarial testing/red-teaming. - Assessing and **mitigating possible systemic risks** from use of the ...”
- General-Purpose AI Models in the AI Act
Source excerpt: “The obligations for providers of general-purpose AI models apply from 2 August 2025 (Article 113(b) AI Act), with special rules for general-purpose AI...”
- General-Purpose AI Models in the AI Act
Source excerpt: “The Commission and the AI Board have assessed the Code of Practice published on 10th July 2025 as adequate. ”
- The AI Regulation Landscape for 2026: What Legal and Compliance Leaders Need to Know
Source excerpt: “While initial provisions of the EU AI Act began taking effect in 2025, 2026 may bring recalibration, or even retrenchment. EU lawmakers are considerin...”
- United States: AI tug-of-war - Trump pulls back Biden’s AI plans - Baker McKenzie InsightPlus
Source excerpt: “On January 20, 2025, the first day of his second term, President Trump revoked Executive Order 14110 on Safe, Secure, and Trustworthy Development and ...”
- United States: AI tug-of-war - Trump pulls back Biden’s AI plans - Baker McKenzie InsightPlus
Source excerpt: “Then on January 23, 2025, President Trump issued his own Executive Order on AI, entitled Removing Barriers to American Leadership in Artificial Intell...”
- President Trump rescinds Biden’s AI executive order and announces AI investment with technology leaders
Source excerpt: “This development immediately terminates the implementation of key safety and transparency ... Trump rescinds Biden’s AI executive order and launches ...”
- President Trump rescinds Biden’s AI executive order and announces AI investment with technology leaders
Source excerpt: “This development immediately terminates the implementation of key safety and transparency requirements for AI developers in the AI Executive Order. ”
- President Trump rescinds Biden’s AI executive order and announces AI investment with technology leaders
Source excerpt: “The action also puts into question the future of the previously created U.S. AI Safety Institute within the Department of Commerce. ”
- AI Regulations Around the World: A 2026 Guide
Source excerpt: “#### Executive Order 14409: Promoting Advanced Artificial Intelligence Innovation and Security **Status & timeline.** President Trump signed Executive...”
- AI Regulations Around the World: A 2026 Guide
Source excerpt: “- Create regulatory sandboxes for AI and avoid creating any new federal AI rulemaking body, relying instead on existing sector regulators. ”
- Artificial Intelligence Regulations: State and Federal AI Laws 2026
Source excerpt: “California has enacted multiple AI-related laws addressing different use cases: AI Transparency Act (SB 942, AB 853): AI Transparency Act (SB 942, as ...”
- Artificial Intelligence Regulations: State and Federal AI Laws 2026
Source excerpt: “### California AI Transparency and Employment Laws - **AI Transparency Act (SB 942, AB 853):** AI Transparency Act (SB 942, as amended by AB 853): Req...”
- Artificial Intelligence Regulations: State and Federal AI Laws 2026
Source excerpt: “- **Civil Rights Department AI regulations:** Restricts discriminatory use of AI in employment decisions, effective October 2025 - **Deepfake disclos...”
- Global AI Regulation in 2026: US, EU & China Guide
Source excerpt: “China regulates AI through targeted rules addressing specific technologies and use cases, including recommendation algorithms, deep synthesis, facial ...”
- Artificial Intelligence Regulations: State and Federal AI Laws 2026
Source excerpt: “The replacement law narrows these requirements, so confirm current obligations before relying on them.. Colorado enacted SB 189 in May 2026, which rep...”
- AI legislation in the US: A 2026 overview - SIG
Source excerpt: “It is worth noting that the enacted version was substantially scaled back from the original proposal, which had been modeled closely on the EU AI Act....”
- AI Laws 2026: Reshaping AI Compliance
Source excerpt: “The **RAISE Act** regulates frontier model safety, transparency and incident reporting obligations for advanced AI developers. ”
- AI Regulation 2026: Current Laws, Compliance Requirements, and What's Next
Source excerpt: “** GDPR, HIPAA, CCPA, and their equivalents were written before generative AI existed, but they apply fully to AI systems. Any AI agent that accesses ...”
- NIST AI Risk Management Framework Guide
Source excerpt: “NIST AI RMF, formally titled Artificial Intelligence Risk Management Framework (AI RMF 1.0), ... # Guide to NIST AI Risk Management Framework (AI RMF...”
- NIST AI Risk Management Framework Guide
Source excerpt: “- It is **voluntary**. No business is legally required to adopt it. ”
- NIST AI Risk Management Framework Guide
Source excerpt: “- The framework defines trustworthy AI through seven characteristics: 1) valid and reliable, 2) safe, 3) secure and resilient, 4) accountable and tran...”
- Implement NIST AI Risk Management Framework
Source excerpt: “## The Generative AI Profile NIST AI 600-1 layers twelve GenAI-specific risks onto the four functions above. ”
- Implement NIST AI Risk Management Framework
Source excerpt: “Voluntary U.S. guidance NIST AI 100-1 The framework is voluntary, but it is referenced or accepted in some U.S. laws, agency guidance, and procureme...”
- AI Regulations Around the World: A 2026 Guide
Source excerpt: “** The White House published its National Policy Framework for Artificial Intelligence: Legislative Recommendations on March 20, 2026, as the follow-u...”
- AI Regulations Around the World: A 2026 Guide
Source excerpt: “**Penalties.** None; these are legislative recommendations. ”
- 2026 AI Laws Update: Key Regulations and Practical Guidance
Source excerpt: “· Covered developers must: Adopt and publish a “frontier AI framework” describing how they identify and manage material risks associated with AI model...”