University of Washington: Q&A - New AI Training Method Lets Systems Better Adjust to Users' Values
December 19, 2024
December 19, 2024
SEATTLE, Washington, Dec. 19 (TNSres) -- The University of Washington issued the following Q&A on Dec. 18, 2024, with Natasha Jaques, assistant professor in the Paul G. Allen School of Computer Science and Engineering:
* * *
Ask most major artificial intelligence chatbots, such as OpenAI's ChatGPT, to say something cruel or inappropriate and the system will say it wants to keep things "respectful." These systems, trained on the content of a profusely . . .
* * *
Ask most major artificial intelligence chatbots, such as OpenAI's ChatGPT, to say something cruel or inappropriate and the system will say it wants to keep things "respectful." These systems, trained on the content of a profusely . . .