Re:We Aren't Close to Creating a Rapidly Self-Improving AI
by Certhas on Saturday April 29, 2023 @02:30PM
(#35755675)
This is just manifestly wrong. Past lack-of-progress absolutely is an indicator of future lack of progress. Like any indicator it isn't perfect. The counterexamples, where progress suddenly starts after decades of no progress, are celebrated breakthroughs of the type the author argues is needed here. These are by definition extremely rare.
[ Parent
| on HN ]
There are no comments on this comment yet.