4 ms·

This is scary. There will be inevitable hallucinations, but this time they might land innocent people in jail.

• sixothree 2y ago
  On top of that, there is the issue of training data. We know police are often less than truthful or complete in their reports. Will these tools follow suit.
• danielmarkbruce 2y ago
  You assume current reports are better. That's probably a bad assumption. So many current fears around AI are based on the fact they might produce an error, as though the status quo is error free. It's not in most fields.
• superfrank 2y ago
  I have an acquaintance who is in a position at a police force that got a very early preview of this. With me working in tech, he wanted my opinion on it and hallucinations were my first concern. I'm not sure if it's still the case, but at that time the way the feature was set up it would add a few sentences of incorrect information into the report intentionally and the report would not be able to be submitted if that information was still in the report. I have no idea if that's how it works now or how well that works, but it seems like the team building this is aware of that possibility and at least taking some steps to try and prevent that from happening.
  • rurp 2y ago
    > it would add a few sentences of incorrect information into the report intentionally and the report would not be able to be submitted if that information was still in the report. This is an interesting safety feature but it would be better if the report submission went through and then resulted in disciplinary measures for the officer who submitted it. Simply blocking the submission makes it too easy to keep submitting the report with small edits in between, while never carefully verifying the full report.
• echoangle 2y ago
  Hopefully the person responsible for writing it reads the generated report and signs that it is accurate.
• johnisgood 2y ago
  Working as intended.