ML's suitability for parallelism mostly comes from the fact that bigger models/state spaces tend to be more powerful. In cases where there no benefit in scaling

ML's suitability for parallelism mostly comes from the fact that bigger models/state spaces tend to be more powerful. In cases where there no benefit in scaling the problem horizontally, you probably won't have much luck scaling the implementation horizontally either.