Roman Yampolskiy on Objections to AI Safety
Future of Life Institute Podcast
00:00
The Upsides of Open Source Language Models
The open sourcing of models or the perhaps the leaked weights of a model from from META have become a large area of concern. Academic research leads to capability at least as much as to safety usually more so yes you learn how to better manipulate errors in that neural network which means now the system can self improve fast and remove its own errors. There seem to be thought experiments you can run which give you information without any improvement in capability anything where you actually working with a model you can even have accidental discoveries.
Transcript
Play full episode
