(Though it’s possible that what lies at the end will be a powerful-enough non-AGI AI tool that it’ll make it very easy for the frontier labs to then use it to R&D an actual AGI, or take over the world, or whatever. This is a subtly different cluster of scenarios, though.)
Why do you say that? They seem functionally very similar, leaving aside that that second possibility suggests we’re in for a terrifying fast takeoff, for the reasons outlined above about lumpiness of innovations.
Why do you say that? They seem functionally very similar, leaving aside that that second possibility suggests we’re in for a terrifying fast takeoff, for the reasons outlined above about lumpiness of innovations.