I thought about it for a while and decided, nah. It’s easy to dismiss offhand that these people are just stupid, but if you think about it a bit longer, these people aren’t just stupid, they are really, really, really, stupid. The model architecture is bad, the data is poisoned, improvement is O(data^2), and performance gains stalled in 2024. No AGI in sight or even a viable method of getting there.
They looked at these charts and thought, the others spent a lot of money, then made a lot of money. If we spend a LOT of money, we’ll make EVEN MORE!
Why don’t we put up the WeWork or MoviePass charts hmmm?
















I feel as though you probably want sand for this…
Also, scipy/MPL, R or GNU Octave if you want pretty things.
If you’re on a chalkboard you can just draw the 2D shape and then use topographical contours.