In short
AI tools help our writers work faster, but people decide what we publish and are responsible for every word. A named writer and an editor check every article, and every fact is verified against a source.
Where we use AI
- Finding, gathering and summarising research and documentation.
- Spotting trending stories (our news radar suggests stories; editors decide).
- Drafting and restructuring passages, which writers then rewrite and check.
- Suggesting headlines, summaries and alternative wording.
- Checking grammar, clarity, consistency and broken links.
- Creating illustrations for some articles, which are labelled as illustrations.
Where we never use AI
- Inventing quotes, sources, statistics, testimonials or user reviews.
- Simulating hands-on testing or describing experiences we did not have.
- Creating fictional writers or bylines.
- Producing images presented as real photographs, or images of real people doing things they did not do.
- Making final editorial decisions, such as review scores and verdicts.
- Processing readers’ personal information or confidential material.
Human oversight
Every article has a named writer responsible for it and is reviewed by an editor before publication. AI output is treated like an unverified tip: useful, but not a source. Facts, figures, quotes and claims are checked against original sources. See our fact-checking policy.
AI-generated images
Some articles use illustrations made with AI image tools. They represent a topic, not a real scene, and we do not use them where they could mislead readers about real events, documents or people. Real product screenshots in hands-on reviews are our own.
Disclosure
When AI played a significant role in creating a piece of content beyond research and editing support, such as an AI-generated data visualisation or image, we say so on the page.
Changes to this policy
AI tools are changing quickly, and so will our use of them. We will update this page when our practices change.