I found this:
Days to receive chronic dose for increase cancer risk of 1 in a 1,000
432 (at 100 CPM)
86 (at 500 CPM)
But it's from a prepper site that doesn't cite their own sources.
I found this: https://www.energy.gov/ehss/articles/doe-ionizing-radiation-...
Which uses rem instead of cpm. An on-line converter of unknown quality says 300 cpm is 500 rem, and the pdf from the .gov site says 500 rem is "death probable in 2-3 weeks", but I think that chart is saying that's whole body & no therapy. Where this is probably mostly hair that can be just cut off totally let alone washed, and so the elevated exposure is probably both low and short duration, and medical therapy (whatever that means, if any in this case) on top.
I can't tell, could be the same as just visting a country with a slightly higher background that isn't a problem for anyone, to dead in a month. Leaning towards no problem just because of the short time and apparently mostly external and removable source.
However, it's not nothing either. It's maybe no problem for this person only because they avoided ingesting the water and the water was very quickly washed off and presumably their hair was cut off and all clothes etc removed as fast as possible. It's clearly at least "rather hot" and you can't just play in it and have prolonged exposure and ingestion. It doesn't seem to be "basically zero".
Assuming this website was used, it looks like it does a naive multiplication by 5/3, which seems... simplistic? The rest of the page doesn't exactly fill me with confidence either. No indication of how the conversion factor was derived and there's a bunch of links to other CPM -> <radiation-related unit> calculators. On top of that, the landing page for the root domain boasts about AI capabilities and their AI page prominently features "Elevate Your Content Creation" and "Generate high-quality AI content with ease!"
[0]: https://www.inayam.co/unit-converter/radioactivity/counts_pe...
It's worth noting that humans are typically radioactive to the level of 3 kBq, or 3000 disintegrations per second, so if I ever realised I had 300 cpm of radiation on my skin as measured by a device that is sensitive to alpha, beta, and gamma, I probably would just shrug and wash it off. Where it might be a problem is if I am dealing with only alpha and beta isotopes, and I'm getting 300 cpm on a gamma-sensitive detector, meaning that the _secondary radiation alone_ is 300 cpm.
(Realistically, I and the radiation safety officer overseeing whatever I was doing would be in serious trouble and have a ton of paperwork, but I just mean it in the abstract)
I mean, 10 grams of potassium has ~300 Bq (that is 300 disintegrations per second) of radiation, so I think I should be able to get my hair far more radioactive than 300 cpm on a beta-sensitive geiger counter if I just slather myself in low-sodium salt from the grocery salt. The salt might be bad for my scalp, I don't know, but the radiation is fine. My point here, though, is that I don't know what equipment the 300 cpm is measured with, what the thresholds are and what the window material is, and that can change things greatly, but my non-professional opinion as the wrong kind of doctor is that it's...probably not a big deal.
We've actually used KCl as a low-level radiation source before, and we joked that when the experiment is done we can just take it home and use it to season dinner.
The linked report doesn't say how radioactive his hair is or give any indication of whether the person in question is threatened by this reading. Could be bad, could be nothing, we just know it is higher than normal.
EDIT The report below it seems to literally be "nothing interesting happened". The thresholds here for something to be reportable are very low. Frankly I don't know why this story is upvoted so much but I'm not about to make a bigger deal about it than one sentence.