The end users’ data IS NOT used to train GPT models by default. The uploaded document is not saved into any permanent server either.

OpenBots Documents can process structured and unstructured documents like scanned documents.

Yes, although the accuracy could be lower, and in some cases, may not work, the new 2.1 upgrade reads handwritten documents if the letters are easily readable. Human-in-the-loop may be necessary in some cases.

Yes, Internally OpenBots documents uses Google, AWS and Microsoft OCR Services.

OpenBots Documents now uses modern AI that typically gives users an accuracy rate of over 95% in most cases. This is at par with Human-Level accuracy. However, this is subjective as accuracy may change based on multiple factors including but not limited to the quality of the scan, the complexity of structures like tables, etc.

We recommend you run a few sample Documents in our secured trial environment to check accuracy. All our documents are fully secured and are not used to train any AI. Documents can be purged immediately or automatically wiped off in 30days.

Yes, OpenBots Documents has a very effective Human In The Loop module also called Validation Station.