It's a very strange algorithm completely different from most compression methods, that's what makes it interesting IMO. But it's probably not realistically competitive, since one needs to do more after the burrows-wheeler transform, and all that "more" has been ridiculously more optimized in zstd and other modern compression methods. Compressing bwt-transformed data is easier, but that doesn't mean it's easy to further than what's easy.
I saw bellard neural net compression benchmarks. If you have time and performant system, based on the semantic of the data to compress, neural net seems tough to beat.
It is a matter of good compromises: speed and efficacy. I even wonder if LZMA2 is worth replacing bzip2.
