Human are still the primary decision point in these 911 calls. However, we’ve seen how often humans accept AI prompts, even when they don’t align with training.
What quality controls did they implement?
Also, what data is the going to the AI company? Hopefully non-voice identifying info is scrubbed.
Bethan
TFW the example you use as a nightmare case of bad application of language technology has actually been implement (and in the works since 2019) in your own city.
buttondown.com/maiht3k/arch...
w/ @deccamuldowney.bsky.social
