How ChatGPT Ruined Christmas

This week’s episode of Approximately Correct explores just how quickly the world of artificial intelligence can change — even for those who are working in it.

Alona Fyshe does double duty as both co-host and guest as she shares the story of a TED talk she was working on in late 2022 about the methods that researchers use to evaluate Large Language Models. Little did she know, there was something very big on the horizon that would change how people thought about AI, forcing her back to the drawing board with only weeks left before her talk.

Listen to Alona talk about how ChatGPT changed the public’s perception of what AI language models can do, instantly making most previous benchmarks seem out of date, and the excitement and intensity of working in a fast-moving field like AI.

You can see the final version of Alona’s talk here.

Approximately Correct: An AI Podcast from Amii is hosted by Alona Fyshe and Scott Lilwall. It is produced by Lynda Vang, and Keely Booth, with video production by Chris Onciul and social media support by Jen Tomski.