Two studies looked at a combined 647 covid-predicting AIs and found that NONE were suitable for clinical use (despite some being probably already in clinical use).
https://twitter.com/joshspero/status/1421365417558282241
"Many unwittingly used a data set that contained chest scans of children who did not have covid as their examples of what non-covid cases looked like. But as a result, the AIs learned to identify kids, not covid."
"Because patients scanned while lying down were more likely to be seriously ill, the AI learned wrongly to predict serious covid risk from a person’s position."
Are these flawed AI models being marketed to hospitals and used on patients? Probably. 
The fact that we discover such widespread problems with AI models as soon as we do independent tests raises some serious questions.
Such as "what other AI products haven't been tested?"
Such as "what other AI products haven't been tested?"
just going to leave this here
https://twitter.com/hatr/status/1361756449802768387?s=20
• • •
Missing some Tweet in this thread? You can try to
force a refresh
