Thanks, this is indeed the case. the US_DRIVER_LICENSE confidence is 0.01 and the demo doesn't put any threshold on the response. We're working on fixing the demo.
Hey, I'm one of the maintainers of Presidio OSS.
We built this tool mainly for Microsoft's strategic customers and decided to open-source it for others to use. No hidden agenda here.
The engine has two main advantages: It's easily expandable and customizable, and it works well at scale. We do know of organizations who are very close to production with it.
Every organization has their own requirements for PII entities, many of them specific to the org itself, so the engine allows a developer to easily add support for new PII entities using code, regex or black-lists.
As for the productization aspect, we did some performance tests and are confident with taking it to production. Single instance cluster with a medium size machine has a ~65ms response time for a 100 word sentence. Using better machines lowers the response time to ~24ms for 100 words and 150ms for a 1,000 words input.
The current service uses regex for known patterns and Spacy for named entity recognition (person names, places etc.). Users often built custom ML models to detect new types of entities.
Presidio is free, completely transparent, and fully customizable. Feel free to use it and let us know what you think.