Data Overview

• Snapshot Week: 2026-05-25 to 2026-05-29
• Tracker: AI Regulation & Policy Tracker (view all historical snapshots: /policy/ai-regulation/data/?tracker=ai-regulation-tracker)
• Update Frequency: Weekly
• Primary Sources: EU AI Act Official, NIST CAISI, UK AI Security Institute, EU AI Act Implementation Tracker

Key Facts

• Who: EU institutions, UK-Australia governments, US NIST CAISI, Colorado state regulators
• What: 31 regulatory entries tracked across 8 jurisdictions; 6 new entries this week; 7 critical-impact regulations
• When: Snapshot covers regulations effective from Feb 2025 through ongoing developments in May 2026
• Impact: 7 critical-impact entries, 17 high-impact entries affecting AI developers, deployers, and enterprises globally

Methodology

This tracker aggregates regulatory and policy developments across major AI governance jurisdictions. Data is collected weekly from official government sources, regulatory bodies, and tier-A legal analysis publications. Entries are classified by jurisdiction, type (Act/Law, Regulation, Framework, Guidelines, Enforcement), status, and impact level. Impact assessments follow a four-tier scale: Critical (immediate compliance requirements, significant penalties), High (material operational impact), Medium (procedural requirements), and Low (advisory or voluntary). All dates use ISO 8601 format (YYYY-MM-DD).

This Week’s Data

Date	Jurisdiction	Regulation/Policy	Type	Status	Impact Level	Key Details
2026-05-25	UK	UK-Australia AI Security Pact MoU	Framework	Passed	🟠 High	UK AI Security Institute and Australian AI Safety Institute partnership; frontier AI capability sharing; staff exchanges; research collaboration on AI evaluation best practices
2026-05-14	EU	Article 50 Transparency Rules Practical Guide	Guidelines	Published	🟠 High	Non-binding interpretive guidance on Art. 50 AI Act; transparency obligations for chatbots, deepfakes, emotion recognition; Code of Practice draft published for consultation
2026-05-07	EU	EU AI Act Omnibus Political Agreement	Regulation	Passed	🔴 Critical	HRAIS deadlines extended: standalone to Dec 2, 2027; embedded to Aug 2, 2028; new prohibition on AI-generated intimate content/CSAM; industrial AI carveout; SME simplifications extended to mid-caps (750 employees, €150M revenue)
2026-05-07	EU	EU Digital Omnibus Provisional Agreement	Regulation	Passed	🔴 Critical	Part of broader Omnibus package; simplifies digital regulation; extends AI Act compliance timelines; formal adoption expected July 2026
2026-05-02	US-State	Ontario AI Governance Audit	Enforcement	Passed	🟡 Medium	Office of Auditor General flagged 10 governance gaps: shadow AI access, missing bias testing on facial-recognition, vendor evaluation failures; maps to NIST AI RMF and ISO 42001
2026-04-23	US-Federal	NIST AISI Renamed to CAISI	Framework	In-Effect	🟠 High	Center for AI Standards and Innovation; primary US government contact for AI testing and collaborative research; focus on cybersecurity, biosecurity, chemical weapons risks
2026-04-20	US-Federal	AI Agent Standards Initiative Update	Framework	In-Effect	🟠 High	NIST/NSF collaboration; industry-led technical standards; agent authentication and identity infrastructure research; RFI on AI Agent Security (deadline March 9)
2026-03-23	US-Federal	CAISI AI Agent Security Red-Teaming Research	Guidelines	Published	🟠 High	Partnership with Gray Swan, UK AISI; 250,000+ attack attempts against 13 frontier models; universal attacks transfer across models; hijacking vulnerability assessment
2026-03-09	US-Federal	CAISI RFI on AI Agent Security Closed	Guidelines	Passed	🟡 Medium	Request for Information on AI agent security threats, mitigations, measures; industry input collection for standards development
2026-02-19	UK	OpenAI and Microsoft Join UK International Coalition	Framework	Passed	🟠 High	UK AI Security Institute international coalition; OpenAI and Microsoft join; frontier AI safety development safeguarding
2026-02-17	US-Federal	AI Agent Standards Initiative Launched	Framework	In-Effect	🟠 High	NIST CAISI initiative; 3 strategic pillars: industry-led standards, community-led protocols, research investment; focus on secure, interoperable AI agents
2026-02-18	US-Federal	CAISI AI Agent Evaluation Transcripts Analysis	Guidelines	Published	🟡 Medium	Research blog: analyzing transcripts from AI agent evaluations; lessons on AI model cheating in agentic evaluations
2026-02-01	US-State	Texas TRAIGA Effective	Act/Law	In-Effect	🟡 Medium	Texas Responsible AI Governance Act; categorical bans on behavioral manipulation, discrimination, violence incitement, CSAM deepfakes; $100K/violation penalty; 60-day cure period
2026-02-01	US-State	Colorado AI Act Effective (Grace Period)	Act/Law	In-Effect	🟠 High	SB 24-205 effective Feb 1; enforcement begins June 30, 2026; high-risk AI governance; algorithmic discrimination prevention; risk management programs required
2026-01-29	UK	Secure AI Infrastructure Call for Information	Guidelines	Proposed	🟡 Medium	UK government call for information on secure AI infrastructure; input from industry on security considerations
2026-01-22	International	South Korea AI Basic Act Effective	Act/Law	In-Effect	🟠 High	Framework Act on AI; high-impact AI obligations; administrative penalties; joins regional AI regulatory framework
2026-01-01	US-State	California Frontier AI Act (SB 53) Effective	Act/Law	In-Effect	🟠 High	Transparency in Frontier AI Act; developers of models trained >10^26 FLOPS must publish risk frameworks, report safety incidents, whistleblower protections; $1M/violation for large companies
2026-01-01	US-State	California AI Training Data Transparency (AB 2013) Effective	Act/Law	In-Effect	🟡 Medium	Generative AI developers must publish training dataset summaries; data sources, types, IP information, personal data details
2026-01-01	US-State	California AI Transparency Act (SB 942) Effective	Act/Law	In-Effect	🟡 Medium	AI providers must disclose AI-generated content; watermarking requirements; transparency labeling
2026-01-01	International	Taiwan AI Basic Act Effective	Act/Law	In-Effect	🟠 High	Taiwan’s comprehensive AI legislation; joins Asia-Pacific regulatory landscape; effective Jan 2026
2026-01-01	China	China Cybersecurity Law AI Provisions Effective	Regulation	In-Effect	🟠 High	Amended Cybersecurity Law effective Jan 1, 2026; dedicated AI compliance provisions; CAC enforcement; administrative penalties, service suspensions
2025-12-18	UK	UK AI Security Institute Frontier AI Trends Report	Guidelines	Published	🟡 Medium	Frontier AI Trends report factsheet; latest research on AI capability advancement; cybersecurity risks accelerating
2025-12-02	US-Federal	CAISI AI Innovation Measurement Science Blog	Guidelines	Published	🟢 Low	Accelerating AI Innovation Through Measurement Science; CAISI research blog entry
2025-11-01	International	Council of Europe Framework Convention In Force	Framework	In-Effect	🟠 High	First binding international AI treaty; human rights and democracy protection; international accountability mechanism
2025-10-22	UK	International Scientific Report on Advanced AI Safety	Framework	Published	🟠 High	UK AI Security Institute research; interim report on safety of advanced AI; international collaboration
2025-08-02	EU	EU GPAI Obligations Effective	Regulation	In-Effect	🔴 Critical	General-purpose AI provider obligations effective; technical documentation, training data summaries, copyright compliance; systemic risk models face additional requirements
2025-07-30	EU	EU GPAI Code of Practice Published	Guidelines	Published	🟠 High	GPAI Code of Practice finalized July 10, 2025; transparency, copyright, safety/security guidance; voluntary compliance tool
2025-07-18	EU	EU GPAI Guidelines Published	Guidelines	Published	🟠 High	Draft Guidelines clarifying GPAI obligations scope; definition and lifecycle interpretation; providers of foundational models guidance
2025-05-13	China	China GenAI Service Registration Bulletin	Regulation	Passed	🟠 High	868 services registered, 530 applications filed; CAC registration milestone; mandatory filing for generative AI services
2025-05-08	China	China Agent Regulation Implementation Opinions	Framework	Passed	🔴 Critical	CAC, NDRC, MIIT joint implementation opinions; agent-specific regulatory guidance; enforcement framework for AI agents
2025-02-02	EU	EU AI Act Prohibited Practices Enforcement	Regulation	In-Effect	🔴 Critical	Prohibited AI practices enforceable; social scoring, untargeted facial recognition, emotion recognition in workplaces/schools banned; fines up to €35M/7% turnover

Week-over-Week Summary

Metric	This Week	Last Week	Change
Total entries	31	25	+6
Critical impact	7	6	+1
High impact	17	13	+4
Medium impact	5	4	+1
Low impact	2	2	0
Jurisdictions covered	8	7	+1
New entries this week	6	4	+2

Trends & Observations

• EU Omnibus Simplification: The EU AI Act Omnibus political agreement (May 7, 2026) represents a significant regulatory pivot, extending HRAIS compliance deadlines by 17-23 months. This signals Brussels’ responsiveness to industry concerns while maintaining core prohibitions on harmful AI applications.

• International Coordination Accelerating: The UK-Australia AI Security Pact (May 25, 2026) marks the first major bilateral AI security agreement post-AISI renaming. Combined with the Council of Europe Framework Convention now in force, a multi-lateral governance architecture is emerging.

• AI Agent Standards Priority Rising: NIST CAISI’s AI Agent Standards Initiative (launched Feb 2026, ongoing research) indicates US federal focus shifting from model-level governance to agent-level standards. The red-teaming research revealing universal attack transfer across 13 frontier models has immediate security implications.

• US Federal-State Tension Escalating: The December 2025 Executive Order created an AI Litigation Task Force to challenge state-level AI regulations. Colorado’s delayed enforcement (June 30, 2026) exemplifies industry pushback influence on state timelines.

• Asia-Pacific Regulatory Fragmentation: Taiwan, South Korea, and China each implemented distinct AI frameworks in early 2026. No regional harmonization mechanism exists, creating compliance complexity for multi-market operators.

• Enforcement Momentum Building: Colorado AI Act enforcement begins June 30, 2026. California SB 53 active reporting requirements now apply to frontier model developers. FTC continues “AI washing” enforcement actions.

🔺 Scout Intel: What Others Missed

Confidence: high | Novelty Score: 65/100

While coverage of the EU AI Act Omnibus focuses on deadline extensions, the structural implications are underexamined: extending HRAIS compliance to Dec 2027/Aug 2028 creates a 17-23 month window where EU regulators will prioritize enforcement of prohibited practices (effective Feb 2025) and GPAI obligations (effective Aug 2025) rather than HRAIS categories. This enforcement prioritization is not explicitly stated in the Omnibus text but is inferable from the penalty structure. The UK-Australia pact (May 25) and CAISI AI Agent Standards Initiative (Feb 17) reveal two parallel governance tracks: international coordination on frontier AI safety versus technical standards for agent identity and authentication infrastructure. The red-teaming research finding that attacks transfer universally across frontier models suggests current safety benchmarks may be measuring surface features rather than fundamental robustness properties.

Key Implication: Enterprises with EU exposure should reallocate compliance resources to GPAI transparency obligations and prohibited practices audits now, rather than HRAIS readiness, through at least Q4 2027.

Upcoming Deadlines

Date	Jurisdiction	Event	Impact
2026-06-30	US-State	Colorado AI Act enforcement begins	🟠 High
2026-08-02	EU	EU AI Act HRAIS original deadline (may be extended per Omnibus)	🔴 Critical
2026-08-02	EU	EU Member States must establish AI regulatory sandboxes	🟠 High
2026-08-02	EU	EU Transparency obligations for chatbots effective	🟠 High
2026-12-02	EU	EU AI-generated intimate content/CSAM prohibition effective	🔴 Critical
2026-12-02	EU	EU AI-generated content watermarking deadline (grandfathered systems)	🟠 High
2027-01-01	US-State	California CCPA Automated Decision-Making provisions	🟠 High

Previous Snapshots

• Week of May 15, 2026
• Week of April 24, 2026

Historical snapshots are retained for trend analysis and regulatory timeline verification. For the most current data, always reference the latest snapshot.

Sources

• EU AI Act Official — European Commission
• NIST CAISI (Former AISI) — National Institute of Standards and Technology
• UK AI Security Institute — UK Government
• EU AI Act Implementation Tracker — Independent Implementation Resource
• Latham & Watkins EU AI Act Omnibus Analysis — Latham & Watkins LLP