{
  "id": 3204329,
  "title": "Met Office finds most of us still won't trust an AI weather forecast",
  "url": "https://urgent.news/2026/08/25/met-office-finds-most-of-us-still-wont-trust-an-ai-weather-forecast",
  "topic": "ai",
  "section": "AI",
  "published": "2026-08-25T06:00:00.000Z",
  "source": {
    "name": "TechRadar",
    "slug": "techradar",
    "url": "https://www.techradar.com/pro/met-office-finds-most-of-us-still-wont-trust-an-ai-weather-forecast"
  },
  "original_language": "en",
  "account": "A new survey from the Met Office reveals that the majority of UK adults remain skeptical about artificial intelligence (AI) weather forecasts. Only 11% of the 6,000 participants expressed \"very high\" confidence in AI-based forecasting techniques, in stark contrast to 30% who trusted the more established numerical and physics-based methods. The Met Office has been utilizing machine learning techniques in weather prediction for a few years, but this survey indicates that the public is still more comfortable with traditional forecasting methods.\n\nThe research, published in a study titled \"Artificial Intelligence for the Earth Systems,\" highlights the \"confidence gap\" between AI and traditional forecasting approaches. While 87.7% of respondents felt confident in weather reports produced using numerical weather prediction (NWP), only 49.4% were open to machine learning weather prediction (MLWP). Dr Edward Pope, the lead author and Met Office Science Fellow, emphasized the importance of clearly demonstrating the benefits of AI-based models to the public in order to bridge this gap.\n\nThe Met Office has been evaluating AI models for their potential to augment existing forecasting methods, but the confidence gap could hinder their wider adoption. Dr Pope stated that \"The gap in confidence and perceived accuracy highlighted in the paper demonstrates the need to clearly and transparently demonstrate the value of new approaches in ways that matter to people.\" Professor Kirstine Dale, the Met Office's Chief AI Officer, added that the research will guide future developments in AI weather modeling, ensuring that any changes are thoroughly evaluated and validated before implementation.",
  "summary": "Survey finds low confidence in weather prediction systems based on machine learning, with the majority overwhelmingly in favour of traditional forecasting.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}