There's a lot of (warranted imo) skepticism there too. Im sorry I can't find the citation but there was a Japanese paper out this year that claimed the ml post

There's a lot of (warranted imo) skepticism there too. Im sorry I can't find the citation but there was a Japanese paper out this year that claimed the ml post processing of the EHT data produces a qualitatively similar image given random data.