They wouldn't let me just refund the one customer. I had to prove the other 25 were delivered, so I did that. Then I had to prove they actually got iPods, so I linked them to the reviewed transactions that showed people were happy with the iPods. So then I had to prove my identity to eBay. They wanted my license, so I did that. Then a utility so I sent my light bill. Then my phone bill. Then my natural gas bill. Then my lease. Then they asked for my passport, which I did not have at that time. Suddenly nothing could be done without the passport, and they'd keep the money for 180 days and then mail me a check. Except then they said they were keeping the money anyway because I didn't provide a passport. Five grand, gone.

From that day on I'd have multiple eBay accounts at all times, spread things out around them so that when (not if, when) they would freeze an account for a review, I'd just cancel everything, refund whatever purchase was in review, and never use that account again. I learned that the review process is just a delay process to make you think there was progress when you'd already been banned but since you thought there was a chance, you'd keep busy in the no-go queue for however long you interacted with the bot before you gave up rather than calling support.

For most of computing history this has been the case, too!

I'll be honest the questions you're quoting WERE simplistic. Is the data sent off site? In 2026? Yes. I tried to give a cheeky answer, sorry you were offended, but yes, in 2026 virtually everyone runs EMRs running off prem.

Then you asked if the data went through an LLM in a post where we were talking about AI transcription tools. At this point I had to assume you didn't know that an AI transcription tool was an LLM, why else would one ask? If you knew the transcriber was an LLM, then why would you ask if the data went through an LLM?

> > Obviously, you can't pricess data you can't access

> Again, it reads to me like you think the questions are simplistic.

They were. However I tried them like they were genuine questions and tried to answer. There have been a thousand times in my career where a person asked a nonsense question like "can I access the data if it's offline?" While the answer is obvious in hindsight, and sometimes people feel embarrassed asking a question like that, avoiding embarrassment isn't a reason to not answer the question. So I said that obviously they can access the data but the risk is in how the data is used, not if it can be accessed, and pointed out the data isn't stored in a permanent form but is processed ephemerally.

I again apologize if I seemed rude. In return I'd ask for an explanation. Why did you ask if the data went offsite when we were talking about AI scribes? Why did you ask if the data was processed by an LLM?

> If there was a diagnosis or treatment I felt others knowing about would compromise me I would avoid bringing it up to a medical professional or seeking treatment. I'm certain there are people who avoid mental health services, for example, for exactly that reason.

I'm very sorry you feel this way and I hope you can find a way to have more trust in your doctors. I assure you data privacy is incredibly important and not something we screw around with. Breaches happen, but it's not the rule.