AI threats in the wild: The current state of prompt injections on the websecurity.googleblog.com4 points·ahoog42··0 commentsOpen articleSaveView on HN