Some states do it this way. Other states do it all electronically (fewer now than in the past, thankfully). Other states do it all on paper and do the counting with offline counting machines, then spot check some precincts at random. Some do it by mail entirely on paper.
And that's the big reason why this line of inquiry is nonsense. The entire country showed a huge shift to the right, not just the swing states or the states that are more vulnerable. That's 51 entirely separate election systems that you'd have to manipulate, make sure public information about the election matches exactly, and also not go so far that any independent exit polls show anything fishy either. The scale of conspiracy to do it in even one state, make no mistakes, and have no one leak is hard to believe. Doing it across the entire country? You're going to need a lot more than "I feel like the numbers are fishy" to be convincing. The conservatives were wrong when they said 2020 was rigged, and anyone saying 2024 was rigged is equally wrong.
Thinking through a problem yourself, or taking an idea and putting it into words, is like exercise for the brain. You may think you understand a thing from reading or hearing about it, but it's only when you do it for yourself that you discover what you really know and what you don't. It's the difference between learning what a square root actually is and how to press the square root button on the calculator. It's the difference between learning to drive and learning to turn on self-driving mode. Even if the outcome is the same, the learning experience is day and night.
Once you understand a concept well enough, then using an LLM to get some busy work done or just get a starting point that you can improve isn't all that bad, much like using a calculator after learning pen-and-paper division, but trying to use one while learning is almost certain to hurt your understanding, even if the LLM doesn't outright make a bunch of stuff up.