Jsevillamol comments on [Prediction] We are in an Algorithmic Overhang, Part 2

[Jsevillamol]

If I’m wrong and the core learning algorithm(s) of human beings is too complicated to write in a handful of scientific papers then superintelligence will not be built by 2121.

Note that even if it is complicated it might be an attractor, so that a not-yet-AGI meta-learning algorithm might stumble upon it.