Data centres and consumer bills. The White House says major tech firms have agreed to a “Ratepayer Protection Pledge” intended to prevent AI data-centre energy costs being pushed onto consumers, alongside grid-related actions.
Reuters reports China’s new five-year plan puts AI across the economy with an ‘AI+’ action plan and an open-source emphasis alongside heavy investment in computing and strategic technologies.
The Office for National Statistics reports new “real-time indicators” outputs that include public views and social trends touching on artificial intelligence, positioning AI as a measurable public-impact issue alongside cost-of-living and daily life concerns.
Regulation
East Dunbartonshire Council publishes an Artificial Intelligence Policy approved on 5 March 2026, setting a local public-sector framework for AI use and oversight. It shows operational policy is moving into day-to-day service delivery, where procurement controls, accountability lines, and transparency duties must be made auditable.
The European Commission publishes updated guidelines on the ethical use of AI and data in teaching and learning, framed as practical support for education stakeholders and ethical-legal compliance.
The Council of Europe announces approval of the HUDERIA Model and COBRA resources, with a consolidated publication made available the same day. Impact assessment methodologies are becoming more standardised, strengthening expectations for structured risk identification and mitigation evidence.
Cases
Supreme Court of the United States denied certiorari in Thaler v Perlmutter (No. 25-449), leaving intact the position that purely AI-generated works are not registrable without human authorship under current US law. Content provenance and human contribution records become a compliance control for IP risk and downstream licensing.
Academia
arXiv lists a 5 March 2026 submission proposing a governance approach for “reliable agentic AI” in a WebGIS development context, explicitly treating governance as a reliability mechanism rather than a post-hoc add-on. This matters for AI governance because it reinforces the need for lifecycle controls, documentation, and testable reliability claims when systems act with higher autonomy.
Events
Data Governance & AI Governance Conference Europe (London) — 23–27 March 2026.
RegTech Annual Conference (City of London) — 26 March 2026.
ICLR 2026 (Rio de Janeiro) — 23–27 April 2026.
WeRobot 2026 — 24–25 April 2026.
Takeaway
Contract design, local public-sector AI policies, and education ethics guidance are converging into a more practical governance toolkit. The compliance gap is increasingly filled by auditable controls, documented restrictions, and impact assessment discipline that can be shown to regulators, users, and procurement counterparts.
Sources: The White House, Reuters, Financial Times, Office for National Statistics, East Dunbartonshire Council, European Commission, Council of Europe, arXiv, WeRobot, ICLR, Kempitlaw, IRMUK