Earth LinC — station Foresight note / 2026 →

The question, asked plainly

AI for Earth

By 2026 every one of those phrases is ordinary. The frontier moves in release cycles; Earth keeps the slowest clock in the room. The missing product is the link.

20262030203520402050

01 What becomes common

Eight things that stop being remarkable.

None of these are predictions any more — they are the floor. Which means none of them are an advantage. The scarce thing in 2026 is coherence between them.

01

AI for AI

Models training, evaluating and repairing models. Self-improvement as infrastructure.

02

Multi-AI

Fleets of agents negotiating, delegating, disagreeing. Orchestration is the new UI.

03

AI library

Weights, tools, prompts and evals as public shelving — versioned, forkable, citable.

04

AI for Earth

Forecasting, remote sensing, materials search. Science at machine tempo.

05

AI for sustainability

Emissions accounting, grid dispatch, circular logistics — mostly reporting, rarely deciding.

06

AI protocol

Shared calling conventions between agents, tools and data. Plumbing, quietly decisive.

07

AI for content

Infinite supply. Provenance, not production, becomes the valuable part.

08

AI with on-chain

Verifiable claims, machine payments, auditable history for what an agent did.

02 Why Earth is behind

The frontier is measured in quarters. Earth is measured in decades.

Sustainability didn't fall behind because it lacked models. It fell behind because its feedback loop is slower than any funding cycle, its data is fragmented across ministries and supply chains, and its wins are avoided losses — invisible by design.

Meanwhile the frontier itself became a physical load. Compute is now an energy story, a water story, a siting story. AI is no longer only a tool pointed at the planet; it is a tenant on it.

415 TWh · data centres, 2024

About 1.5% of global electricity, on a path toward roughly 945 TWh by 2030 in the IEA's base case.[1]

4.4% of US electricity, 2023

176 TWh in US data centres, with a projected range of 6.7–12% of national demand by 2028.[2]

17% of SDG targets on track

Against a 2030 deadline. Nearly half show minimal progress; a third have stalled or reversed.[3]

Intelligence got cheap. Consequence didn't.

03 How you link it

Four layers, in order. Skip one and the chain breaks.

A link between AI and Earth is not a campaign. It is a stack — each layer only useful because the one beneath it is trustworthy.

Layer 01 · measure

Make the planet machine-readable

Satellite, sensor and inventory data fused into continuous emissions, water and land estimates instead of annual self-reported PDFs. Independent measurement is what turns a pledge into a number.[4]

Layer 02 · protocol

One vocabulary for agents and Earth data

Shared schemas and tool-calling conventions so an agent in a logistics firm and a model in a grid operator mean the same thing by "a tonne", "a kilowatt-hour", "a hectare". Interoperability is the multiplier the frontier already gave itself — Earth data never got it.[5]

Layer 03 · ledger

Provenance that outlives the vendor

On-chain or otherwise cryptographic records for what was measured, by whom, with which method — plus the same for what an AI system did. Content credentials and audit trails, applied to physical claims.[6]

Layer 04 · incentive

Wire it to money and rules

Disclosure regimes, procurement standards, energy-efficiency codes and market prices that read layer 03 directly. Without this, the first three layers are a very expensive dashboard.[7]

04 Horizons

Two futures per decade. The fork is the link.

2030

The accounting decade

Emissions must fall roughly 42% below 2019 levels this decade for a 1.5°C-consistent path, while current policies point closer to 3°C. In the same window AI electricity demand roughly doubles. Whoever makes measurement cheap and continuous sets the terms for everything after.

Linked — every model, dataset and data centre reports energy, water and siting as a native field. Unlinked — sustainability stays a report; compute growth is argued about with estimates.[1][8]
2035

The agent-operations decade

Multi-agent systems move from drafting to dispatching: grids, water networks, freight, building stock. The binding question is no longer capability but authority — what an autonomous system is permitted to change in the physical world, and who can audit it afterwards.

Linked — agents operate physical systems under logged, revocable mandates. Unlinked — optimisation for cost quietly overrides optimisation for carbon.[5][7]
2040

The materials decade

AI-accelerated discovery matures into deployed chemistry and manufacturing: catalysts, storage, cement, alloys, recycling. This is where machine intelligence stops being a reporting tool and becomes the reason a hard-to-abate sector is abatable at all.

Linked — discovery pipelines are ranked by lifecycle impact, not just performance. Unlinked — we invent brilliant materials with unexamined footprints.[9]
2050

The settlement decade

Net-zero commitments come due, and the record is read back. By then the archive matters more than the ambition: which claims were verifiable, which models were audited, which decisions had a signature. 2024 already sat about 1.55°C above pre-industrial — the ledger is being written now, whether or not we keep it.

Linked — a continuous, contestable, machine-readable planetary record. Unlinked — a generation of unfalsifiable claims.[10]

05 What's next

AI for X — the X's that are still open.

"AI for X" stops being a slogan when X is something with a physical bill attached. These are the frontiers where the work is unglamorous, the data is missing, and the leverage is largest.

AI for grids

Forecasting, flexibility and interconnection queues — the bottleneck between clean supply and real demand.

AI for materials

Cement, steel, storage, catalysts. The sectors no software has yet decarbonised.

AI for water

Basin-level allocation, leak detection, and the cooling footprint of AI itself.

AI for soil

Sub-field agronomy and carbon in the ground, verified rather than assumed.

AI for repair

Diagnosis, spare-part matching, reuse markets — intelligence pointed at longevity.

AI for provenance

Where a thing came from and who touched it, for atoms as well as for content.

AI for consent

Machine-readable permission over land, data, labour and likeness. Governance as an API.

AI for slow time

Systems that can hold a fifty-year objective without being retuned every quarter. The hardest one.

06 Sources

Every number above, traceable to a primary document.

Cited by publisher, exact title and year so you can retrieve the original and read past the summary — and so this page stays honest as the figures are revised.

  1. International Energy Agency — Energy and AI, 2025Global data-centre electricity ~415 TWh in 2024 (~1.5% of demand); ~945 TWh base case by 2030 · supports §02, §04
  2. Lawrence Berkeley National Laboratory — 2024 United States Data Center Energy Usage Report, 2024US data centres 176 TWh (4.4% of US electricity) in 2023; 6.7–12% projected by 2028 · supports §02
  3. United Nations — The Sustainable Development Goals Report, 2024Roughly 17% of SDG targets on track for 2030 · supports §02
  4. Climate TRACE — Global Greenhouse Gas Emissions Inventory, ongoingIndependent satellite- and sensor-derived, facility-level emissions estimates · supports §03 layer 01
  5. Open Geospatial Consortium & ISO/IEC — Geospatial and AI management standards (ISO/IEC 42001:2023)Interoperable Earth-observation schemas and AI management systems · supports §03 layer 02, §04
  6. Coalition for Content Provenance and Authenticity (C2PA) — Content Credentials SpecificationCryptographic provenance model transferable to physical and environmental claims · supports §03 layer 03
  7. European Union — Corporate Sustainability Reporting Directive (2022/2464) and AI Act (Regulation 2024/1689)Mandatory disclosure and risk-tiered AI obligations — the incentive layer in law · supports §03 layer 04, §04
  8. UN Environment Programme — Emissions Gap Report, 2024~42% emissions cut below 2019 by 2030 for 1.5°C; current policies ≈3°C · supports §04 / 2030
  9. Merchant et al. — Scaling deep learning for materials discovery, Nature, 2023 (GNoME)AI-predicted stable crystals at scale; basis of the materials horizon · supports §04 / 2040
  10. World Meteorological Organization — State of the Global Climate 2024, 20252024 approximately 1.55°C above the 1850–1900 baseline · supports §04 / 2050
  11. IPCC — AR6 Synthesis Report: Climate Change 2023Decadal carbon budgets and mitigation pathways underpinning all four horizons

No live hyperlinks by design: report URLs rot and mirrors mislead. Search the exact title with its publisher and year, and you will land on the primary document — the version you can quote.

Read the primary documents