I’d imagine a lot of this data on workflows was collected and stored for use in training models to replace the workers. The part on handwriting unique introduct

I’d imagine a lot of this data on workflows was collected and stored for use in training models to replace the workers. The part on handwriting unique introductions or judging trustworthiness seem like low hanging fruit to automate now.