Skip to main content

Chatbots are so gullible, they’ll take directions from hackers

“Prompt injection” is a major risk to large language models and the chatbots they power. Here’s how the attack works, examples and potential fallout.


https://wapo.st/40ixi1b

Popular posts from this blog

Ask Help Desk: Stop auto-correct from ducking up

From perplexing auto-corrections to the best way to call 911, we are answering your biggest tech questions. https://wapo.st/3nqPcfV
Read more

U.S. charges North Korean man with cyberattacks on NASA

Rim Jong Hyok is accused of using ransomware to attack computer systems at American hospitals, generating money to pay for hacking NASA and U.S. military bases. https://www.washingtonpost.com/national-security/2024/07/25/north-korea-hacking-indictment/?utm_source=dlvr.it&utm_medium=blogger
Read more