Chatsh EBT's Performance Changes Over Time

3min Snip

00:00

Play full episode

Summary

Transcript

Episode notes

Researchers from Stanford and UC Berkeley put out a paper having tested it. They found big changes including some large decreases in some problem solving tasks. Between March and June, GPT4 became much less willing to answer sensitive questions. When it came to coding, they also found a significant degradation in how directly executable the code produced by both GPT4 and GPT3.5 actually was.

A new research paper from Stanford and Berkeley says that yes, indeed ChatGPT has gotten worse. However others are calling the methodology into question. NLW explores the debate. Before that on The Brief, OpenAI shuts down AI detection tool; Sec. Blinken writes AI op-ed; Egypt using AI to restore Mummies. Today's Sponsor: Supermanage - AI for 1-on-1's - https://supermanage.ai/breakdown ABOUT THE AI BREAKDOWN The AI Breakdown helps you understand the most important news and discussions in AI. Subscribe to The AI Breakdown newsletter: https://theaibreakdown.beehiiv.com/subscribe Subscribe to The AI Breakdown on YouTube: https://www.youtube.com/@TheAIBreakdown Join the community: bit.ly/aibreakdown Learn more: http://breakdown.network/