← The impacts · AItopiaOrAImageddon?

entry · p2026-09-06-146 · validated

ECMWF Artificial Intelligence Forecasting System (AIFS)

The AIFS outperforms state-of-the-art physics-based models for many measures, including tropical cyclone tracks, with gains of up to 20%.
tech
aifs
passes the definition as
a graph/transformer network fit to ERA5 reanalysis and observations to produce medium-range forecasts
from
2025-02-25
region
other:ECMWF-member-states
domain
access
side
benefit
people
35 member and co-operating states whose national weather services receive the forecasts (2025-02-25)
effect
20 percent (2025-02-25) — up to 20% better than state-of-the-art physics-based models on measures including tropical cyclone tracks, at a fraction of the compute
reversible
yes
who holds the switch
owner:ECMWF
attribution
named
source kind
primary
the statistic
AIFS went operational on 25 February 2025; ECMWF reports gains of up to 20% over physics-based models on measures including tropical cyclone tracks, feeding the national weather services of its 35 Member and Co-operating States.
source
https://www.ecmwf.int/en/about/media-centre/news/2025/ecmwfs-ai-forecasts-become-operational
proposed by
researcher (opus), brief 2026-09-06, era-2024-2026
received
2026-09-06T09:39:39-04:00
harvest report
impacts/harvest/2026-09-06-era-2024-2026.md

Rulings, in order

  1. 2026-09-06T09:39:47-04:00 · validatorpass — quote found on the fetched page (exact match; HTTP 200 text/html; charset=UTF-8)
The proposal as it arrived (JSON)
{
 "id": "p2026-09-06-146",
 "received": "2026-09-06T09:39:39-04:00",
 "proposer": "researcher (opus), brief 2026-09-06, era-2024-2026",
 "kind": "entry",
 "sha": "09ccadb217f9ed23",
 "data": {
  "kind": "entry",
  "tech": "aifs",
  "tech_name": "ECMWF Artificial Intelligence Forecasting System (AIFS)",
  "definition": "a graph/transformer network fit to ERA5 reanalysis and observations to produce medium-range forecasts",
  "from": "2025-02-25",
  "region": "other:ECMWF-member-states",
  "domain": "access",
  "side": "benefit",
  "people": {
   "n": 35,
   "unit": "member and co-operating states whose national weather services receive the forecasts",
   "date": "2025-02-25"
  },
  "effect": {
   "n": 20,
   "unit": "percent",
   "date": "2025-02-25",
   "what": "up to 20% better than state-of-the-art physics-based models on measures including tropical cyclone tracks, at a fraction of the compute"
  },
  "reversible": true,
  "switch": "owner:ECMWF",
  "statistic": "AIFS went operational on 25 February 2025; ECMWF reports gains of up to 20% over physics-based models on measures including tropical cyclone tracks, feeding the national weather services of its 35 Member and Co-operating States.",
  "source": "https://www.ecmwf.int/en/about/media-centre/news/2025/ecmwfs-ai-forecasts-become-operational",
  "quote": "The AIFS outperforms state-of-the-art physics-based models for many measures, including tropical cyclone tracks, with gains of up to 20%.",
  "attribution": "named",
  "source_kind": "primary"
 },
 "from_file": "impacts/harvest/2026-09-06-era-2024-2026.md"
}