alt Hacker NewsHow confident are you that this knowledge was not part of the training data? Was there no stackoverflow questions/replies with it, no tech forum posts, private knowledge bases, etc?
Not trying to diminish its results, just one should always assume that LLMs have a rough memory on pretty much the whole of the internet/human knowledge. Google itself was very impressive back then in how it managed to dig out stuff interesting me (though it's no longer good at finding a single article with almost exact keywords...), and what makes LLMs especially great is that they combine that with some surface level transformation to make that information fit the current, particular need.
Do you think AlphaGo is regurgitating human gameplay? No it’s not: it’s learning an optimal policy based on self play. That is essentially what you’re seeing with agents. People have a very misguided understanding of the training process and the implications of RL in verifiable domains. That’s why coding agents will certainly reach superhuman performance. Straw/steel man depending on what you believe: “But they won’t be able to understand systems! But a good spec IS programming!” also a bad take: agents absolutely can interact with humans, interpret vague deseridata, fill in the gaps, ask for direction. You are not going to need to write a spec the same way you need to today. It will be exactly like interacting with a very good programmer in EVERY sense of the word