Evidence Checker
What we are, what we aren't, and why we built something different.
Health claims are everywhere. On social media, in wellness blogs, from well-meaning friends — and increasingly, from AI assistants. The problem isn't access to information. It's knowing which information to trust.
Most people, when they ask an AI about a health claim, get an answer that sounds authoritative. What they don't get is an answer that has been filtered through the standards scientists actually use to evaluate evidence. A single small study gets weighted the same as a large randomised controlled trial. An association gets reported as a cause. A finding in mice gets applied to humans. A fringe position gets equal billing with the scientific consensus.
This isn't a failure of AI — it's a failure of the question. Without very specific parameters around what constitutes good evidence, and how to weight different types of research against each other, even a capable AI tool will return confident-sounding answers that don't reflect the full picture.
CITE was built to solve that. We've developed a structured approach to evidence evaluation — so you don't need to know how to interrogate a scientific literature base yourself. We do it for you, the way a careful scientist would.
In science, not all evidence is equal. A well-designed randomised controlled trial (RCT) tells us far more than an anecdote or an observational study. A meta-analysis — which pools findings from multiple high-quality studies — is more reliable than any single trial on its own. Systematic reviews that synthesise the available literature carry more weight than individual expert opinion.
CITE applies this logic to every claim it evaluates. Rather than simply retrieving information, it works through a structured evidence hierarchy — prioritising the research that carries the most scientific weight, and being explicit about where the evidence is weak, mixed, or absent.
We've also built in specific checks for the most common ways health science gets misrepresented:
The result is an evaluation that reflects not just what research exists, but how much confidence the science actually warrants.
CITE is an evidence-checking tool. It's designed to help you understand what the current scientific literature says about a health claim, and how strong that evidence actually is. Think of it as a rigorous starting point for understanding — not a replacement for professional advice.
CITE is not:
If you're making an important health decision, please talk to a qualified professional who knows your personal circumstances.
Every claim CITE evaluates receives one of four verdicts, based on the overall weight of the available evidence.
The claim is supported by strong, consistent scientific evidence. Multiple well-designed studies — or meta-analyses synthesising the research — point in the same direction with reasonable confidence.
The evidence is mixed, partial, or preliminary. There may be support for the claim, but significant caveats apply — inconsistent findings across studies, limitations in study quality, small sample sizes, or important nuances that affect how the evidence applies in practice.
The claim is contradicted by evidence, or the available research does not support it. A red verdict doesn't always mean the claim is entirely impossible — it means the scientific evidence, on balance, doesn't back it up.
There isn't enough quality evidence to reach a verdict. The research may be too limited, too early-stage, or too conflicting to draw a reliable conclusion. This is an honest answer, not a failure — some claims genuinely haven't been well studied yet.
We've built CITE to be as rigorous as possible, but no tool is perfect. It's important to understand where the limits are.
We're committed to improving CITE over time. If you receive a result that seems wrong or incomplete, treat it as a prompt to dig deeper — not as a final word.
Disclaimer. CITE is an informational tool only. Nothing on this platform constitutes medical advice, diagnosis, or treatment recommendation. The information provided is for general educational purposes and reflects the state of the scientific literature as understood at the time of evaluation. Individual health circumstances vary significantly — always consult a qualified healthcare professional before making decisions about your health.