Editorial standards

How we rank sources

1. Systematic reviews and peer-reviewed studies
2. Regulator documents (TGA, FDA, EMA)
3. Official company documents (treated as claims only)
4. Reputable journalism
5. Everything else

Press releases and social posts show what was claimed, never that it's true.

How our ratings work

Each breakthrough gets one of four ratings: Approved & in use, In clinical trials, Early research or Announced only. The rating can never be higher than the evidence stage of the claims that support it. When the evidence is mixed, we follow the weakest key claim and explain why.

How we handle numbers

We state study size and type, whether results held up in a separate group of patients, and who funded it where known. We give absolute numbers beside relative ones, and explain accuracy terms in plain words.

Approval wording

FDA "cleared", "granted" and "approved" mean different things, and we use the exact word. Approval in one country isn't approval in another. Every page has a Where is it approved? block covering the US (FDA), EU (CE mark/EMA), UK (MHRA), Australia (TGA) and others where relevant. If we haven't verified a status, we say Unknown.

Stable links

Every claim has a permanent ID and anchor (for example /tracker/ai-mammogram-screening/#c2). Page addresses never change; if one must, it gets a permanent redirect.

Reviews

Every page is re-checked at least every 90 days. Pages past their review date show a "Being reviewed" note.

Corrections

If we get something wrong, we fix it, add a dated note to the page and its change history, and list it on our Corrections page. Spotted an error? Use our contact form.