How do you judge claims of shared remote-viewing hits?

I’d start with the protocol, not the most impressive examples. Were viewers kept unaware of the target and the other sessions? Were their notes and sketches recorded before anyone saw the image? Could the target have been chosen or changed after the sessions?

Then decide in advance how results will be scored. Compare each description with the actual target and with decoy images using the same criteria, preferably by evaluators who do not know which image is correct. A timestamp can help establish when a record was made, but it cannot by itself show that the description matched well.

Several people making similar descriptions sounds intriguing, but it is hard to assess from a highlight reel. If a group does many sessions across a large pool of possible images, some overlap or memorable-looking matches may happen by chance. I’d want to see the full set of attempts, including misses, not just selected examples.

A good next step would be a simple test that anyone can audit: randomly select targets, keep viewers separate, lock each record before feedback, and have independent scorers compare the results against the target plus decoys. Repeat it with the rules fixed. That would make it easier to tell whether the pattern holds up beyond individual examples.

Different people may still disagree about what counts as a hit, so showing the raw descriptions and the scoring method matters as much as showing the best matches.