Most claim checking in public produces two outcomes: true or debunked. Real evidence rarely cooperates, and a field this young produces a great many questions where the honest answer is that nobody has done the work properly.
What unknown means
Not that the claim is doubtful, and not a polite refusal. It means the evidence required to answer does not exist, is not accessible, or does not address the question being asked.
Four situations produce it: no human study exists; human studies exist but not in a relevant population; the studies conflict without a clear reason; or the underlying evidence is not public, which happens with proprietary formulations.
Unsupported and unknown are different
Unsupported means we looked and the evidence does not carry the claim. Unknown means the evidence to decide has not been produced. A reader should treat those differently, and a checker that has only one bucket has made the decision for them.
Why the third state has to be structural
A system with two outputs will always convert its uncertainty into whichever answer is cheaper to defend. In health that is reliably the negative, because a wrongly encouraged reader complains and a wrongly discouraged one simply goes away.
So unknown cannot be a matter of good intentions. It has to be a state the process can actually output, or it will be quietly rounded away every time under pressure to look decisive.
Why longevity produces so many of them
The outcome of interest takes decades to observe. Trials that long are expensive and rare. Much of the commercial money funds product development rather than independent verification. And the field moves faster than the evidence, so most compounds are being discussed years before any human trial reports.
None of that is a scandal. It is the ordinary state of a young field, and it means a high rate of honest unknowns is what a truthful checker should produce.
What we publish alongside an unknown
An unknown verdict is not the end of an entry. It names what evidence exists, what specifically is missing, whether any trial is underway and when it reports, and what result would change the verdict. That last one is the part that makes it useful rather than merely cautious.
How to read one
An unknown is not a recommendation in either direction. It means anyone telling you the answer confidently is going beyond the evidence, including anyone telling you confidently that it does not work. Longevity Claim Verification exists to draw that line clearly rather than to have an opinion about everything.