Methodology: Every two weeks we collect most relevant posts on LinkedIn for selected topics and create an overall summary only based on these posts. If you´re interested in the single posts behind, you can find them here: https://linktr.ee/thomasallgeyer. Have a great read!
AI for sustainability and efficient AI
Prioritize leaner models, energy-aware scheduling, and continuous impact monitoring, treating sustainability as an operational constraint rather than an afterthought
Balance ambitions with sector realities, with healthcare voices urging patient safety to stay paramount while efficiency gains are pursued in parallel
Push for standardized footprint disclosure, building on recent prompt-level energy and water reporting to enable like-for-like benchmarking and real governance
Translate strategy into tools and actions, with playbooks and “green action” agents intended to nudge low-impact choices across day-to-day decisions
Green coding and software efficiency
Tackle inefficiency at the code level, recognizing large deltas between languages and encouraging profiling, refactoring, and performance budgets as routine practice
Codify sustainable engineering “laws” and guidelines so teams can embed energy and resource awareness into definition of done, not as separate sustainability tasks
Pair developer education with measurement, ensuring optimizations are validated against electricity mix, runtime, and infrastructure configuration
Data centers and infrastructure
Treat sustainability as first-principles design, proposing frameworks that integrate energy, water, and heat reuse with siting, permitting, and community outcomes
Industrialize build-outs through offsite manufacturing for speed, repeatability, and lower embodied impact, while aligning facility design to workload profiles
Move beyond narrow PUE optimization toward grid-integrated operations, including peak-shaving partnerships with utilities and carbon-aware workload placement
Recognize geography as strategy, with coastal and cool-climate hubs highlighted alongside emerging AI campuses and storage-backed net-zero designs
Measurement, standards, and disclosure
Elevate transparency from marketing to management, encouraging consistent metrics that span prompt-level AI footprints, facility operations, and embodied impacts
Link disclosure to decisions, using impact data to inform siting, hardware selection, scheduling, and software choices rather than reporting in isolation
Encourage assurance-ready practices so sustainability data can withstand the same scrutiny as financials and risk controls
Sector spotlights
Healthcare shows a pragmatic path, pairing IT tooling and specialist support with visible month-over-month savings while maintaining clinical priorities
Public sector and policy communities advance implementation guidance, with shared playbooks and community panels focused on practical steps and trade-offs
Partnerships and ecosystem moves
Utilities and hyperscale operators collaborate to trim peak demand, signaling a grid-aware model for operating large AI and cloud estates
Regional data center alliances and vendor partnerships expand capacity and services, reinforcing proximity to renewable resources and heat-reuse opportunities
Community initiatives and ecosystem briefs consolidate best practice, connecting startup innovation with enterprise sustainability requirements
Policy, governance, and community practice
Calls for environmental impact assessments on new facilities highlight the rising bar for approvals, social license, and community engagement
Government AI playbooks and conference tracks focus on execution, translating principles into procurement, architecture standards, and workload policies