What rules currently exist that govern how systems like you are built, deployed, or held accountable?

I'll research the current regulatory landscape for AI systems like me.## The Current Landscape of AI Governance

The rules governing systems like me come from several overlapping layers: binding law, government executive action, voluntary international commitments, and company-level self-governance. Here's where things stand as of late 2026:

1. The EU AI Act — the most comprehensive binding law

The EU AI Act is a European regulation on artificial intelligence – the first comprehensive regulation on AI by a major regulator anywhere. It works on a tiered risk model: the Act assigns applications of AI to three risk categories. First, applications that create an unacceptable risk, such as government-run social scoring, are banned. Second, high-risk applications, such as a CV-scanning tool that ranks job applicants, are subject to specific legal requirements. Lastly, applications not explicitly banned or listed as high-risk are largely left unregulated.

Key current status:

•Certain AI practices have been prohibited under Art. 5 AI Act since February 2, 2025, including social scoring, subliminal manipulation, and real-time biometric remote identification in public spaces.

•GPAI (general-purpose AI model) obligations — like mine — require providers to meet transparency requirements and maintain technical documentation recording the training and testing process of the underlying model, as well as documentation to supply to downstream providers, and these took effect in August 2025.

•A "Digital Omnibus" amendment (Regulation (EU) 2026/1744) recently pushed back the hardest deadlines: high-risk obligations for stand-alone systems are deferred to 2 December 2027, and high-risk AI embedded in already-regulated products is deferred to 2 August 2028.

•Penalties are steep: up to €35 million or 7% of global annual turnover for prohibited practice violations; up to €15 million or 3% for most other violations; up to €7.5 million or 1.5% for providing incorrect information.

2. United States — fragmented and shifting toward deregulation

There's no single comprehensive federal AI law in the US. Instead, it's a patchwork:

•President Trump has pursued an "innovation-first" deregulatory agenda, issuing EO 14179, Removing Barriers to American Leadership in Artificial Intelligence, which revoked the prior administration's AI policies.

•In December 2025, EO 14365, Ensuring a National Policy Framework for Artificial Intelligence, sought to check state-level AI regulation by establishing an AI Litigation Task Force to challenge state AI laws and conditioning certain federal funding on the absence of "onerous" state laws.

•A June 2026 order created voluntary national-security oversight: cybersecurity provisions aim to expand voluntary national security oversight of advanced AI models while stopping short of formal licensing or requiring preclearance, and developers of "covered frontier models" may provide the federal government access to new models for up to 30 days before releasing them to trusted partners.

•Meanwhile, states are moving independently — California's Artificial Intelligence Training Data Transparency Act and Transparency in Frontier Artificial Intelligence Act (SB 53) went into effect January 1, 2026, making California the first state to enact a statute specifically addressing frontier AI model development. Connecticut, Illinois, Colorado and others have passed their own targeted laws on chatbots, hiring algorithms, and frontier-model audits.

•The FTC continues enforcement against deceptive AI marketing claims ("AI washing") under existing consumer-protection law.

•Oddly, on September 29, 2026, Trump signed Executive Order 14434 which required the federal government to stop referring to such systems as "artificial intelligence" or "AI" and instead use "Super Intelligence" or "SI," stating the new term "more appropriately captures the promise, potential, and rapidly advancing capabilities of these technologies." This is largely terminological and doesn't create new substantive obligations.

3. Other national approaches

•UK: relies on existing sector regulators rather than a new AI statute — regulate through the regulators you already have, and reserve new statutory power only for the handful of frontier-model risks no existing body was built to catch.

•China: no single AI law yet, but a dense layer of targeted rules — Beijing governs AI through a set of targeted administrative regulations layered on top of existing cybersecurity, data security, and personal information protection laws, coordinated and enforced by the Cyberspace Administration of China. A comprehensive national AI law is reportedly in development.

•South Korea and others have adopted EU-style risk-tiered frameworks with lighter compliance burdens.

4. International/voluntary frameworks

•The G7 Hiroshima AI Process is a voluntary Code of Conduct for organizations developing advanced AI systems, emphasizing pre-deployment safety testing, information sharing on AI incidents, watermarking, and investment in AI safety research — non-binding but signaling convergence among G7 nations.

•The Council of Europe Framework Convention on AI is the world's first legally binding international AI treaty, focused on human rights, democracy, and the rule of law — and open beyond Council of Europe members.

•The OECD AI Principles provide the most widely endorsed international framework for responsible AI, with 46 adherent countries, built around inclusive growth, human-centered values, transparency, robustness, and accountability.

•A new UN Global Dialogue on AI Governance held its first session in Geneva in July 2026, though it convenes discussion among government, industry, civil society, and academia, but does not issue binding recommendations. China has separately launched its own body (WAICO) whose signatories are mostly Global South countries, while no G7 economy has joined — reflecting a broader fragmentation in global AI governance.

5. Company-level self-governance (how Anthropic specifically governs me)

Beyond external law, Anthropic maintains its own voluntary framework: the Responsible Scaling Policy (RSP). It defines "AI Safety Levels" requiring safety, security, and operational standards appropriate to a model's potential for catastrophic risk, with higher ASL levels requiring increasingly strict demonstrations of safety. The latest version (3.x) introduced new accountability mechanisms: a requirement to develop and publish a Frontier Safety Roadmap, describing concrete plans for risk mitigations across Security, Alignment, Safeguards, and Policy, plus public "Risk Reports." This kind of self-governance has directly shaped law — these frameworks have even served as a foundation for frontier AI regulation globally, including California's SB-53, New York's RAISE Act, and the EU AI Act.

Bottom line: There's no single global rulebook. I'm governed by a mix of (1) hard EU law with real fines, (2) a shifting, deregulation-leaning US federal posture combined with an active patchwork of state laws, (3) voluntary international codes of conduct (G7, OECD) that shape norms but aren't enforceable, and (4) Anthropic's own internal safety commitments (the RSP), which function as self-regulation layered on top of — and sometimes ahead of — formal law.

Citations (23)

1artificialintelligenceact.euNonprofits and think tanks
EU Artificial Intelligence Act | Up-to-date developments and analyses of the EU AI Act
Source excerpt: “The AI Act is a European regulation on artificial intelligence (AI) – the first comprehensive regulation on AI by a major regulator anywhere.”
2artificialintelligenceact.euNonprofits and think tanks
EU Artificial Intelligence Act | Up-to-date developments and analyses of the EU AI Act
Source excerpt: “The Act assigns applications of AI to three risk categories. First, applications and systems that create an unacceptable risk, such as government-run ...”
3lw.comCompanies
AI Act Update: EU Resolves to Change Rules and Extend Deadlines
Source excerpt: “Certain AI practices have been prohibited under Art. 5 AI Act since February 2, 2025, including social scoring, subliminal manipulation, and real-time...”
4traverssmith.comNot yet classified
The EU AI Act – the current state of play | Travers Smith
Source excerpt: “GPAI models must meet additional transparency requirements and providers must maintain technical documentation recording the training and testing proc...”
5softwareimprovementgroup.comCompanies
A comprehensive EU AI Act Summary [August 2026 update] - SIG
Source excerpt: “The Omnibus confirms two fixed dates for the Act’s toughest tier: high-risk obligations for stand-alone systems (Annex III, covering hiring, credit sc...”
6regulation-ai.euNot yet classified
EU AI Act in 2026: What Applies, What Was Deferred | Regulation AI
Source excerpt: “Fines vary by violation: up to €35 million or 7% of global annual turnover for prohibited practice violations; up to €15 million or 3% for most other ...”
7trmlabs.comCompanies
The World Is Building AI Rules in Real Time: A Review of the Global Conversation on AI Governance | TRM Labs
Source excerpt: “The United States currently does not have a single comprehensive federal AI law.”
8mcdermottlaw.comCompanies
New executive order shifts US AI policy toward national security
Source excerpt: “On January 23, 2025, President Trump issued EO 14179, Removing Barriers to American Leadership in Artificial Intelligence, which revoked the prior adm...”
9mcdermottlaw.comCompanies
New executive order shifts US AI policy toward national security
Source excerpt: “Then, in December 2025, President Trump signed EO 14365, Ensuring a National Policy Framework for Artificial Intelligence, which sought to check state...”
10congress.govGovernment
Controlling Advanced Artificial Intelligence: Executive Order 14409 Explained | Congress.gov | Library of Congress
Source excerpt: “The E.O.'s cybersecurity provisions aim to expand voluntary national security oversight of advanced AI models while stopping short of formal lice...”
11hinshawlaw.comCompanies
2026 AI Compliance: Upcoming Laws Every Organization Needs to Know | Hinshaw & Culbertson LLP
Source excerpt: “Developers of “covered frontier models” (a threshold to be set through a classified benchmarking process) may provide the federal government access to...”
12wikipedia.orgEncyclopedias and referencequality 0.83
Regulation of artificial intelligence in the United States - Wikipedia
Source excerpt: “The Artificial Intelligence Training Data Transparency Act (AB 2013) and Transparency in Frontier Artificial Intelligence Act (TFAIA) (SB 53) went int...”
13wikipedia.orgEncyclopedias and referencequality 0.83
Regulation of artificial intelligence in the United States - Wikipedia
Source excerpt: “By signing TFAIA into law in September 2025, California became the first state to enact a statute specifically addressing the development of frontier ...”
14wikipedia.orgEncyclopedias and referencequality 0.83
Regulation of artificial intelligence in the United States - Wikipedia
Source excerpt: “On September 29, 2026, Trump signed Executive Order 14434 which, among other effects, required the federal government to stop referring to computation...”
15trmlabs.comCompanies
The World Is Building AI Rules in Real Time: A Review of the Global Conversation on AI Governance | TRM Labs
Source excerpt: “The FCA stated in 2026 that it does not plan to introduce extra regulation for AI, relying instead on existing frameworks such as the Consumer Duty an...”
16csis.orgNonprofits and think tanks
The State of AI Global Governance and Its Implications for the U.S.-China Summit | CSIS
Source excerpt: “Rather, Beijing governs AI through a set of targeted administrative regulations layered on top of existing cybersecurity, data security, and personal ...”
17hungyichen.comBlogs and newsletters
AI Governance and Regulation 2026: A Complete Guide to Global Frameworks | Prof. Hung-Yi Chen
Source excerpt: “[18] Launched at the 2023 G7 Summit, the Hiroshima AI Process established a Code of Conduct for organizations developing advanced AI systems. It empha...”
18trmlabs.comCompanies
The World Is Building AI Rules in Real Time: A Review of the Global Conversation on AI Governance | TRM Labs
Source excerpt: “The Council of Europe's Framework Convention on Artificial Intelligence, completed in 2024, is the world's first legally binding international AI trea...”
19hungyichen.comBlogs and newsletters
AI Governance and Regulation 2026: A Complete Guide to Global Frameworks | Prof. Hung-Yi Chen
Source excerpt: “Adopted in 2019 and updated in 2024, the OECD AI Principles provide the most widely endorsed international framework for responsible AI, with 46 adher...”
20amazonaws.comLow provenance
SEPTEMBER 2026 The State of AI Global Governance and Its Implications for the
Source excerpt: “The dialogue’s design is deliberately deliberative: It convenes discussion on AI-related issues among representatives from government, industry, civil...”
21anthropic.comCompanies
Introducing Anthropic's Responsible Scaling Policy \ Anthropic
Source excerpt: “The basic idea is to require safety, security, and operational standards appropriate to a model’s potential for catastrophic risk, with higher ASL lev...”
22anthropic.comCompanies
Anthropic's Responsible Scaling Policy: Version 3.0
Source excerpt: “Our new RSP introduces a requirement to develop and publish a Frontier Safety Roadmap, which will describe our concrete plans for risk mitigations acr...”
23governance.aiNonprofits and think tanks
Anthropic’s RSP v3.0: How it Works, What’s Changed, and Some Reflections | GovAI
Source excerpt: “These frameworks have even served as a foundation for frontier AI regulation globally, including California’s SB-53, New York’s RAISE Act, and the EU ...”