Artificial intelligence is moving from demos into operating systems, public services and physical infrastructure — and the costs are becoming harder to ignore. OpenAI says one enterprise team cut a launch workflow from 243 active role-hours to 77 using Codex and ChatGPT Work; the UK is deploying AI to triage millions of misdirected 101 calls; and a proposed London hyperscale datacentre would consume as much electricity as more than a million UK homes. The practical implication is clear: adoption is accelerating, but governance, resilience and energy planning now determine whether these systems scale responsibly. Read OpenAI’s case study of Stampli’s launch workflow.
What else matters
UK police will use AI to redirect non-emergency calls
A Home Office and National Police Chiefs’ Council system, backed by £1.4m and developed with Vodafone Business, will identify why people call 101 and redirect non-policing queries before they reach a handler. Pilots across 15 forces diverted issues including noise complaints and fly-tipping, with safeguards sending uncertain or urgent cases to police. Read the UK government announcement on the 101 triage system.
London’s proposed AI datacentre exposes the infrastructure trade-off
Planning documents for Digital Reef’s £14.7bn East Havering campus estimate 1.2m tonnes of CO₂e annually at full operation and electricity demand equivalent to more than one million UK households. The project would require major National Grid upgrades by 2033, making power access and carbon accounting central to the UK’s AI buildout. Read the Guardian’s report on the North Ockendon proposal.
EU AI Act enforcement has entered its implementation phase
The European Commission’s framework says the AI Office and national authorities are responsible for implementing, supervising and enforcing the Act from 2 August 2026, including powers over general-purpose AI models. For developers and deployers, the immediate task is translating risk categories, transparency duties and documentation into operating controls. Read the European Commission’s AI Act framework.
The takeaway
The next phase of AI competition will be measured less by model launches alone and more by whether companies and governments can deploy systems inside real constraints: accountable decisions, dependable public-service handoffs and enough clean power to run them.
