Security researchers are jailbreaking large language models to get around safety rules. Things could get much worse.
Read the story here.
Learn more about your ad choices. Visit podcastchoices.com/adchoices