in

Google Eats Rocks, a Win for A.I. Interpretability and Safety Vibe Check

Google Eats Rocks, a Win for A.I. Interpretability and Safety Vibe Check



This week, Google discovered itself in additional turmoil, this time over its new AI Overviews function and a trove of leaked inside paperwork. Then Josh Batson, a researcher on the A.I. startup Anthropic, joins us to elucidate how an experiment that made the chatbot Claude obsessive about the Golden Gate Bridge represents a serious breakthrough in understanding how giant language fashions work. And lastly, we check out current developments in A.I. security, after Casey’s early entry to OpenAI’s new souped-up voice assistant was taken away for security causes.

Guests:

Additional Reading:

“Hard Fork” is hosted by Kevin Roose and Casey Newton and produced by Whitney Jones and Rachel Cohn. The present is edited by Jen Poyant. Engineering by Alyssa Moxley and authentic music by Dan Powell, Elisheba Ittoop, Marion Lozano, Sophia Lanman and Rowan Niemisto Fact-checking by Caitlin Love.

Special due to Paula Szuchman, Pui-Wing Tam, Nell Gallogly, Kate LoPresti and Jeffrey Miranda.

Report

Comments

Express your views here

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Disqus Shortname not set. Please check settings

Written by EGN NEWS DESK

Sarah McLachlan Is Resurfacing

Sarah McLachlan Is Resurfacing

The New ChatGPT Offers a Lesson in A.I. Hype

The New ChatGPT Offers a Lesson in A.I. Hype