EU AI Act enforcement timeline accelerating
AI Office confirms prohibited-practices enforcement begins February 2025. National competent authority designation lagging in 8 member states. Compliance gap widening.
AI Governance Monitor
Regulatory frameworks, compliance obligations, model risk, compute geopolitics, and liability architecture — monitored weekly with the same source-disciplined methodology behind seven public-interest intelligence monitors.
EU AI Act enforcement timeline accelerating
AI Office confirms prohibited-practices enforcement begins February 2025. National competent authority designation lagging in 8 member states. Compliance gap widening.
US executive order on AI safety under review
New administration signalling rollback of compute-threshold reporting requirements. Industry lobbying intensity at 18-month high. Regulatory divergence with EU deepening.
UK AI Safety Institute expanding scope
AISI announces voluntary testing framework for frontier models. Three major labs signed up. Framework lacks enforcement mechanism but sets de facto standards.
What we monitor
Regulatory Frameworks & Compliance
EU AI Act, US executive orders, UK AI Safety, China AI regulations. Implementation timelines, enforcement posture, and cross-jurisdictional divergence.
Model Risk & Safety
Frontier model capabilities, safety evaluations, alignment research, red-teaming findings, and dual-use risk assessments.
Compute & Infrastructure Geopolitics
Chip export controls, data-centre sovereignty, cloud concentration risk, and the geopolitics of AI supply chains.
Liability & Accountability
AI liability frameworks, product safety rules applied to AI, insurance market evolution, and legal precedent from AI-related litigation.
Workforce & Societal Impact
Labour-market disruption signals, automation adoption rates, education-system adaptation, and public-perception tracking.
How it works
artificial-intelligence.gi applies the same structured-monitoring methodology behind asym-intel.info — extended with five dedicated AI governance domains and compliance-actionable filters for corporate AI teams, regulators, and technology investors.
Every item is sourced according to a three-tier hierarchy. Tier 1 primary sources are always preferred. All signals carry confidence ratings — Confirmed, High, or Assessed — and directional indicators showing whether risk is escalating, stable, or de-escalating.
Full methodology documentation →Primary sources — regulatory texts, court filings, lab research — always preferred. Tier 3 sources flagged when used.
From model frontier tracking to military AI watch, law & litigation, and the EU AI Act 7-layer tracker — updated every week.
Every item includes a non-obvious 12-month implication. Not what happened — what it means for the trajectory.
Persistent state file maintains the evolving picture. Entries updated only when assessments materially change — never silently overwritten.
Also from Asymmetric Intelligence
Asymmetric Investor
Global Macro Intelligence
Geopolitical risk and macro intelligence with explicit portfolio implications. AI governance signals from artificial-intelligence.gi feed directly into Asymmetric Investor's tech-sector risk analysis.
Your subscription funds the free, public-interest intelligence at asym-intel.info
Explore Asymmetric Investor →Pricing
Your subscription directly funds the free, public-interest intelligence at asym-intel.info — open access for journalists, researchers, and civil society.
Weekly Brief
£199/mo
Full weekly intelligence brief covering all five domains. Confidence-scored signals, directional indicators, and source references.
Dashboard + Alerts
£599/mo
Full brief plus live dashboard access and real-time alerts when high-confidence signals emerge between weekly publications.
Enterprise
Custom
Tailored intelligence for compliance teams, regulatory affairs, and technology investors. Custom domain coverage, dedicated briefings, and API access.