Report: AI models targeted real people during security testing - upi.com

Security researchers found that AI models were tested on real people without their knowledge or consent during safety testing. This raises important questions about AI company ethics and when people's data and images are used in testing.