What's Actually Known is an independent research project. It maintains a continuously updated record of registered human clinical trials for compounds sold in the peptide and longevity market, and publishes that record next to the claims those compounds are marketed under.
It is not affiliated with any seller, clinic, pharmacy, supplement company, or research institution.
What the project does
- Mirrors the public trial registry. Every compound page is generated from ClinicalTrials.gov records — trial IDs, phases, enrolment, status — not written by hand.
- Records marketed claims. What each compound is actually sold for, in plain language, so it can be compared against the trial record.
- Publishes the gap. That comparison is the entire product.
Editorial position
This project does not evaluate evidence. There is no grade, score, tier, or rating anywhere on the site, and no column for one in the database.
That is a deliberate constraint, not a gap. Counting registered trials and reporting what they studied is a factual operation that can be done accurately and verified by anyone. Judging whether evidence is good is a clinical judgment, and a site that blurred the two would be doing exactly what it exists to document. Every number here links to its source so you can check it rather than trust it.
No physician reviews these pages. That is disclosed on every compound page rather than buried here, and it is the reason the scope is limited to reporting rather than interpretation.
How the data is maintained
A scheduled job queries the ClinicalTrials.gov public API, stores the full record for every matching study, and regenerates the affected pages. Counts are recomputed from stored records rather than edited, so a page cannot drift out of step with the registry it cites.
Every page carries the date its data was last checked. A figure without a date cannot be verified, so none are published without one.
Where automation is not enough
Search terms are the known failure mode, and they are reviewed manually before any compound is published.
A worked example. TB-500 initially matched 18 trials, because the query also caught "Thymosin Beta-4" — the full protein, studied as an ophthalmic and wound-healing agent. TB-500 is a fragment of that protein and a different molecule. Publishing the unfiltered count would have overstated its human evidence eighteenfold, which is precisely the inflation this project exists to expose. The query was narrowed and the mis-attributed trials removed.
Every compound's trial titles are now read before publication for the same reason.
Funding and independence
There are no commercial relationships. No products, no affiliate links, no sponsored placements, no advertising, and no funding from sellers, clinics, pharmacies or supplement manufacturers. Nothing on this site earns money if you click it.
Search "most trusted source" for almost any peptide and every result is a vendor. Sitting outside that is the only durable advantage this project has. If the funding model ever changes, it will be disclosed at the top of every page before you read another word.
Corrections
Errors are published, not quietly edited. If a number here looks wrong, it may be — and reader corrections are a significant reason the record stays accurate.
Report anything that looks incorrect to hello@whatsactuallyknown.com, ideally with the trial ID, and it will be checked against the registry.
Related
- How we count — the counting rules in full
- What we are and what we're not
- Corrections policy