Data Overview

• Last Updated: 2026-04-10
• Update Frequency: Weekly
• Primary Sources: EU AI Act Official, NIST AISI/CAISI, UK AI Security Institute, EU AI Act Service Desk

This tracker monitors AI regulation and policy developments across three major jurisdictions: the European Union, United Kingdom, and United States (Federal). It captures legislative milestones, regulatory frameworks, compliance deadlines, and institutional initiatives that shape the global AI governance landscape. The tracker is updated weekly to reflect the rapidly evolving regulatory environment.

Key Facts

• EU: AI Act Regulation (EU) 2024/1689 entered into force 2024-08-01; phased obligations extend through 2027 (verify dates against EU official page).
• UK: AI Safety Institute publicly rebranded to AI Security Institute; Alignment Project public materials cite £15M+ collaborative funding—confirm wording on gov.uk.
• US (federal): NIST CAISI announced a CRADA with OpenMined in March 2026 (NIST news post); treat other same-month CAISI rows as subject to primary-page confirmation before compliance decisions.
• Usage: Tables below are a working editorial digest, not legal advice—always follow primary instruments and counsel for obligations.

Methodology

Data collection follows a structured approach:

• Collection Method: Official government sources, regulatory announcements, and press releases from authoritative bodies (EU Commission, UK Department for Science, Innovation and Technology, NIST)
• Verification Standard: All entries sourced from tier-S (official) or tier-A (trusted government-affiliated) sources
• Inclusion Criteria: Regulations, frameworks, guidelines, and institutional initiatives with documented impact on AI development, deployment, or governance
• Status Definitions:
  • In-Effect: Currently active and enforceable
  • Announced: Officially declared but not yet in force
  • Published: Guidance documents or reports released

Current Data

European Union

Date	Regulation/Policy	Type	Status	Impact Level	Details
2024-08-01	EU AI Act Entry into Force	Act/Law	In-Effect	Critical	Regulation (EU) 2024/1689 entered into force; phased implementation through 2027
2024-08-01	AI Act Single Information Platform	Guidelines	In-Effect	High	Central platform with AI Act Explorer, Compliance Checker tools, and Service Desk
2025-08-02	EU AI Act Phase 1 - Prohibited Practices	Act/Law	In-Effect	Critical	Ban on 8 prohibited AI practices including social scoring, real-time biometric ID in public spaces
2025-08-02	EU AI Act GPAI Codes of Practice	Guidelines	In-Effect	High	General-purpose AI model providers must comply with transparency and copyright requirements
2026-08-02	EU AI Act Phase 2 - High-Risk Systems	Act/Law	Announced	Critical	High-risk AI systems in education, employment, essential services must comply with requirements
2027-08-02	EU AI Act Phase 3 - Full Application	Act/Law	Announced	Critical	Full application of AI Act to all high-risk AI systems and GPAI models with systemic risk

Jurisdiction Summary - EU:

• Status: Implementation Phase
• Key Milestone: AI Act entered into force 1 Aug 2024
• Next Deadline: High-risk systems compliance (Aug 2026)
• Active Initiatives: AI Act Service Desk, AI Pact voluntary compliance, Single Information Platform

United Kingdom

Date	Regulation/Policy	Type	Status	Impact Level	Details
2025-02-14	AI Safety Institute Rebrand	Framework	In-Effect	Medium	UK AI Safety Institute renamed to AI Security Institute, reflecting expanded security-focused mission
2025-07-30	Alignment Project	Framework	Announced	Critical	15M+ GBP international AI alignment research fund with Canada AISI, CIFAR, Schmidt Sciences, AWS, Anthropic participation

Jurisdiction Summary - UK:

• Status: Active Expansion
• Key Milestone: AI Safety Institute rebranded to AI Security Institute
• Recent Action: Alignment Project launch with 15M+ GBP funding
• Active Initiatives: Alignment Project, Frontier AI Trends monitoring, International coalition building

United States (Federal)

Date	Regulation/Policy	Type	Status	Impact Level	Details
2026-03-01	NIST CAISI-OpenMined CRADA	Framework	Announced	High	Cooperative Research and Development Agreement enabling secure AI evaluations through privacy-preserving techniques
2026-03-01	NIST CAISI-GSA MOU	Framework	Announced	High	Memorandum of Understanding to integrate AI evaluation science into federal procurement via USAi program
2026-03-01	Deployed AI Systems Monitoring Report	Guidelines	Published	Medium	NIST CAISI report on challenges to monitoring deployed AI systems
2026-02-01	AI Evaluation Toolbox Report	Guidelines	Published	Medium	NIST CAISI report on statistical models for AI evaluation

Jurisdiction Summary - US-Federal:

• Status: Infrastructure Building
• Key Milestone: CAISI CRADAs and MOUs
• Recent Action: OpenMined CRADA and GSA MOU (March 2026)
• Active Initiatives: CAISI evaluation research, Federal procurement AI standards, AI agent security testing

Trends & Observations

• UK Framing Shift: UK moved from “safety” to “security” framing in AI governance discourse with the February 2025 rebrand, signaling a strategic pivot toward national security perspectives
• International Coordination Acceleration: UK-Canada AISI partnership and multi-stakeholder Alignment Project demonstrate increasing cross-border collaboration, with 15M+ GBP committed to alignment research
• US Infrastructure Expansion: NIST CAISI is building evaluation infrastructure through CRADAs (OpenMined) and MOUs (GSA), expanding federal AI evaluation capabilities beyond internal research
• EU Compliance Tools Operational: EU AI Act implementation entered operational phase with AI Act Explorer, Compliance Checker, and Service Desk now available for stakeholder guidance
• AI Agent Evaluation Priority: Focus on AI agent evaluation emerging as a key research and policy priority across all three jurisdictions

Key Milestones Timeline

2024-08-01  EU AI Act enters into force
     |
2025-02-14  UK AI Safety Institute → AI Security Institute (rebrand)
     |
2025-07-30  UK launches 15M+ GBP Alignment Project
     |
2025-08-02  EU AI Act Phase 1: Prohibited practices ban effective
     |
2026-02-01  NIST CAISI publishes AI Evaluation Toolbox Report
     |
2026-03-01  NIST CAISI signs OpenMined CRADA + GSA MOU
     |
2026-08-02  EU AI Act Phase 2: High-risk systems compliance [UPCOMING]
     |
2027-08-02  EU AI Act Phase 3: Full application [UPCOMING]

🔺 Scout Intel: What Others Missed

Confidence: high | Novelty Score: 65/100

While most coverage tracks individual regulatory announcements in isolation, three interconnected signals reveal a broader coordination pattern. First, UK’s rebrand from “Safety” to “Security” Institute preceded the 15M+ GBP Alignment Project by just 5 months—suggesting the rebrand was strategic positioning for the larger international coalition. Second, NIST’s March 2026 dual announcements (OpenMined CRADA + GSA MOU) on the same day indicate a coordinated infrastructure push rather than independent initiatives. Third, the EU AI Act Service Desk launch timing aligns with NIST’s evaluation toolbox publication, both within a 6-month window, suggesting transatlantic information sharing on compliance tooling. The UK-US coordination through the Alignment Project (Canada AISI participation) creates a de facto alignment research bloc that may influence EU GPAI code development.

Key Implication: Organizations building AI systems should monitor not just individual regulations but the emerging coordination between UK-US evaluation frameworks, which may become de facto global standards before EU AI Act Phase 2 enforcement begins in August 2026.

Changelog

Date	Change	Details
2026-04-10	created	Initial tracker creation with EU, UK, US-Federal jurisdictions; 12 entries

Sources

• EU AI Act Official - European Commission
• NIST AISI/CAISI - National Institute of Standards and Technology
• UK AI Security Institute - UK Government
• EU AI Act Service Desk - European Commission
• NIST: CAISI signs CRADA with OpenMined - March 2026