AI research projects are not anticipating that the computing power they are going to use in the project will have consequentially increased during the research period.

Where are we exactly on the curve of accelleration to singularity.