Hacker News Reader

> AI is about right 1% of the time

I have an app where people scan their retail/restaurant/cafe receipts and store them. I use Gemini 2.0 for OCR.

It is right more than 99% of the time. I would be making many more errors transcribing the receipts manually.

AIs are a different kind of tools, for a different kind of problems, with a different kind of failure cases.

OCRs are quite different from AI/ML, which the parent comment is referring to. In fact, OCR existed way before the current AI wave.

OCRs are AI/ML in the sense it meant to everyone before 2020.

In case it wasn't clear, I'm using "OCR" in my comment above for "reading and interpreting textual documents provided as images", not as a reference to a specific tech such as Tesseract.

In the app, I am, in fact, using AI/ML tech and I'm saying it's very accurate for my use case, as a counterpoint to the parent comment which says AI works in 1% cases, maybe.

But since you brought it up, this workflow works better than the supposedly "99.999999% right all the time" traditional code (traditional OCR tools).

Hope this clears it up.

Reply on news.ycombinator.com