4 ms·
Introducing Superalignment (July 2023)
- gnabgib 2y agoPreviously on HN [0] (219 points, 11 months ago, 370 comments) Related OpenAI's Long-Term AI Risk Team Has Disbanded [1](86 points, 7 hours ago, 48 comments) [0]: https://news.ycombinator.com/item?id=36603382 https://news.ycombinator.com/item?id=36603382 [1]: https://news.ycombinator.com/item?id=40390831 https://news.ycombinator.com/item?id=40390831
- behnamoh 2y ago> We are dedicating 20% of the compute we’ve secured to date over the next four years to solving the problem of superintelligence alignment. Meanwhile Jan Leike's tweet after his resignation: > Over the past few months my team has been sailing against the wind. Sometimes we were struggling for compute and it was getting harder and harder to get this crucial research done. (https://x.com/janleike/status/1791498182125584843 https://x.com/janleike/status/1791498182125584843)