Over just a few months, ChatGPT went from correctly answering a simple math problem 98% of the time to just 2%, study finds

from what i understand, the big change in chat-gpt4 was that the model could “ask for help” from other tools: for maths, it knew it was a maths problem, transformed it to something a specialised calculation app could do, and then passed it off to that other code to do the actual calculation

same thing for a lot of its new features; it was asking specialised software to do the bits it wasn’t good at

My guess is that it's more a result of overfitting for alignment. Fine-tuning for "safety" (rather, more corporate-friendly outputs).

That is, by focusing on that specific outcome in training the model, they've compromised its ability to give well-"reasoned" "intelligent" sounding answers. A tradeoff between aspects of the model.

It's something that can happen even in simple statistical models. Say you have a scatter plot of data that loosely follows some trend, and you come up with two equations to describe that trend. One is a simple equation that loosely follows it but makes a good general approximation, and the other is a more complicated equation that very tightly fits the existing data. Then you use those two models to predict future data. But you find that the complicated equation is making predictions way off the mark that no longer fit the trend, and the simple one still has a wide error (how far its prediction is from the actual data) but still more or less accurately fits the general trend. In the more complicated equation, you've traded predictive power for explanatory power. It describes the data you originally had but it's not useful for forecasting data that follows.

That's an example of overfitting. It can happen in super-advanced statistical models like GPT, too. Training the "equation" (or as it's been called, spicy autocorrect) to predict outcomes that favor "safety" but losing the model's power to predict accurate "well-reasoned" outcomes.

If that makes any sense.

I'm not a ML researcher or statistician (I just went through a phase in college), so if this is inaccurate I'm open to corrections.

Yep.

Standard VC bullshit.

Burn money providing a lot for nothing to build brand recognition. Then cut free service before bringing out "premium" that at first works better than the original.

Until a bunch of people starting paying and the resources aren't scaled up to match.