The stance is defensible but the article reads like it was written in 2023. Still using the term 'stochastic parrot', in 2026, when century-old mathematical problems are being proven by LLMs almost on a weekly basis doesn't make sense.
The solving of the mathematics problems is largely a marketing stunt. It's a great headline though.
It was also deeply uneconomic, and didn't change the world, or materially advance our understanding. Mostly it's demoralised the mathematicians already working to solve those problems. They're not going to say 'That's great, I can now devote my energies to another problem', they're more likely to look at 10 years of poorly paid intellectually rigorous work and go home and kick the dog. For a headline.
Would these mathematicians have said the same thing if another mathematician with a strong competitive drive has beat them to the timing? And how do these mathematicians think of the proof of the four color theorem?
Comments
The stance is defensible but the article reads like it was written in 2023. Still using the term 'stochastic parrot', in 2026, when century-old mathematical problems are being proven by LLMs almost on a weekly basis doesn't make sense.
The solving of the mathematics problems is largely a marketing stunt. It's a great headline though.
It was also deeply uneconomic, and didn't change the world, or materially advance our understanding. Mostly it's demoralised the mathematicians already working to solve those problems. They're not going to say 'That's great, I can now devote my energies to another problem', they're more likely to look at 10 years of poorly paid intellectually rigorous work and go home and kick the dog. For a headline.
Would these mathematicians have said the same thing if another mathematician with a strong competitive drive has beat them to the timing? And how do these mathematicians think of the proof of the four color theorem?
I suppose we don't have the same definition of 'stochastic parrot'.