How worried—or excited—should we be about AI?
Channels with Peter Kafka
The Limits of Transparency in Language Models
The GPT-4 paper says that they believe it's a matter of safety to not disclose anything about the training data. In order to be able to use this safely, we need to know what's in it. The way they're getting this air's-outs fluency is by taking everything they can possibly get their hands on. That strategy leads to these data sets that are probably too big to document thoroughly.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
