AI chatbots have failed people in crisis. Can that be fixed?
Clinicians and researchers say AI companies need to open up their safety data.
This is a summary aggregated from Ars Technica. Read the complete article on the original site:
Read full article at Ars Technica
