As the underlying Twitter thread points out[1], there's a significant confounding factor here: the sample is from 2019 but might have been processed in 2020, at which point the Chinese government's (public) timeline for COVID's spread admits of the possibility of contamination.

I don't think that completely deflates the evidence here. But as a layperson, I would like to understand the probability of happening to find COVID in an Antarctic soil sample in 2019 versus the probability that someone accidentally contaminated the sample with early samples in early 2020. Put another way: my (lay) intuition doesn't understand why there would be COVID in Antarctic soil, whereas it does understand how contamination might happen in a lab that was actively processing samples of COVID.

[1]: https://twitter.com/jbloom_lab/status/1491297806115807233

particularly the end of README.md:

"On Feb-8-2022, I e-mailed the Chinese authors of the paper to ask about the sample deletion and restoration. They e-mailed back almost immediately. They confirmed what they had told Istvan: they had sequenced the samples with Sangon Biotech (Shanghai) after extracting the DNA in December 2019 from their samples. The suspect that contamination of the samples happened at Sangon Biotech. They deleted the three most contaminated samples from the Sequence Read Archive. They do not know why the samples were then 'un-deleted.'"

(end quote)