We use AI tools. Readers are entitled to know where, and to know that a named human remains responsible for everything we publish.
Where we use AI
- Research assistance and background reading.
- Summarising documents we have obtained.
- Drafting support and editing.
- Translation of source material, always checked against the original.
- Data processing and chart preparation, under our data methodology.
Where we do not
- We do not publish AI-generated text as journalism without human verification and rewriting.
- We do not use AI to generate quotations, sources, statistics or citations.
- We do not use AI to make editorial judgements about what to publish, whom to name, or how to characterise a person or organisation.
- We do not use AI-generated images of real people, places, facilities or events.
Accountability
Every article has a named human author responsible for its accuracy. “The AI produced it” is never an explanation for an error; it is an error by the person who published it. Errors are handled under our corrections policy.
Illustrations
Where an illustration is AI-generated, it is labelled as such.
Before publishing: This is the most common place for a policy to drift from practice. Read each line and confirm it describes what you actually do. If you use AI more heavily in drafting, say so plainly — an honest broader policy is far safer than a narrow one that turns out to be untrue.
This policy sits under our editorial policy, and our wider disclosures are on the transparency page.