Earlier quoted context omitted.
In many businesses, the employee is responsible for inputting most of that. If a LLM can get to 95% accuracy and flag exceptions, the employees (and AP team) would actually have less work and bureaucracy. Though we’ve had a few incidents where employees have submitted AI-generated receipts for reimbursement which is another issue..
Please tell me those are former employees. How can anyone feel confident committing such blatant fraud.
Yeah the ones that were just fraud we termed almost immediately.. there were a few in the middle ground of where they’d lost their receipts (or decided that fabricating a receipt was easier) but we could verify that the expense actually took place from a cc statement. We added an extra approval step for those employees going forward which is annoying for everyone involved so hopefully they’ve learned their lesson there.