"However, these concepts are still in their infancy and are potentially decades away from being used in the real world." Well, maybe. There will need to be new AI architecture concepts. Compact liquid neural networks have significant promise. Yann Lecun's H-JEPA concept is also an encouraging path forward.

The biggest change is easier, letting go of Artifical General Intelligence as the Holy Grail. Humans aren't generally intelligent. We have widely varying natural capabilties and even more widely varying training in special skills. Speciallized intelligent agents are dramatically improving today and will be one of histories greatest value-creating and most disruptive transformations. What they require is high quality, specialized data.