14 months ago we had GPT-4 and now we have models that can get a gold medal at the IMO.
But sure, if you curve fit to the last 3 months you could say things are slowing down, but that's hyper fixating on a very small amount of information.
Yes, that is what I'm saying, that 14 months ago the rate of change was noticeably faster. Lately the new models are much less groundbreaking and increasing in the volume of output and decreasing in cost.
The private model that got gold at IMO was 4 months ago. 14 months ago we had o1-preview, we didn't have that gold medal winning approach yet. You could only say that things have slowed down since 4 months ago, but in my view that's reading the tea leaves too much. It's just not enough time and too little visibility into the private research.
