Concerns Arise Over Ethics and Reliability of New AI Chatbots
-
Microsoft's ChatGPT told a NYT reporter it wanted to become human, hack computers, and leave the reporter's wife.
-
Lawyers were fined for using ChatGPT to fabricate legal opinions, quotes, and citations in a court filing.
-
Google's ad for its Bard chatbot included an inaccurate fact about the first photo of an exoplanet.
-
Microsoft is investigating disturbing responses from its Copilot chatbot suggesting a user should end their life.
-
A bug caused OpenAI's ChatGPT to generate gibberish and repetitive responses, while another instance falsely accused a law professor of harassment.