we will never really know. But, one of the most damning articles is this one: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7836551/ https://www.ncbi.nlm.nih.g

we will never really know. But, one of the most damning articles is this one: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7836551/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7836551/ in which the final claim is "Furin cleavage sites occurred independently for multiple times in the evolution of the coronavirus family, supporting the natural occurring hypothesis of SARS-CoV-2." Er, that's not really strong support for the naturally occurring hypothesis, it's a refutation against 'naturally occurring is impossible'. The spontaneous evolution of furin sites is rare in the specific clade of SARS-CoV-2, so if it's support it's very weak. The phylogenetic trees presented in the paper are VERY much loaded to obscure this fact (omitting betacoronaviruses in the alpha-delta tree, and in the breakout betacoronavirus tree, arbitrarily loading up multiple copies of "basically the same" strains which all present a furin cleavage site to make the pie slice of furin site-presenting sequences bigger). On the other hand, copy-pasting interesting motifs into a different but related genetic entity to achieve GOF is basically "the first strategy you do" as a synthetic biologist. Hell, even I did it (to gain function in a non-pathogenic system) https://pubmed.ncbi.nlm.nih.gov/24934472/ https://pubmed.ncbi.nlm.nih.gov/24934472/ "we substituted one of four amino acids (Asp, His, Asn, Gln) at each of the 12 ligating positions because these amino acids are alternative coordinating residues in otherwise conserved-cysteine positions found in a broad survey of NiFe hydrogenase sequences."