Pedro Domingos Describes AI Progress Cycles
The University of Washington professor outlines a repeating pattern of hype and stalls in AGI efforts.
TLDR
Pedro Domingos, Professor Emeritus of Computer Science at the University of Washington and author of The Master Algorithm, posted on X about the typical progression of AI development. He stated that smart but eccentric individuals first predict imminent AGI. They then achieve notable advances and believe success is near. Progress then halts for a period. A new group of researchers later repeats the same sequence. The post frames this as a generational cycle in the field.
Combined views
6.2K
1 Source, first seen ago
Sources
PD 
Pedro Domingos@pmddomingosHow AI progresses: 1. Some smart but loony people think they're going to achieve AGI in no time. 2. They make some remarkable progress and think they're almost there. 3. Time passes and progress stalls. 4. The next generation comes along and repeats.
- likes: 100
- replies: 15
- bookmarks: 6
- reposts: 1