Piss on the poor arse website strikes again.
It's not a research paper. There doesn't appear to be a link to paper in it. It didn't even meet the threshold of a tiny pilot study (which I said was inadequate). It was also (allegedly) used for idea generation which I specifically called out as not fucking sufficient.
Let's get into the article.
Did you read the fucking article? It's not subtle that it's marketing. It is phrased quite peculiarly when referring to what the LLM and the people using it did, right? Isn't that interesting? Christ, it's not even clear if the LLM was used at all!
The company OpenAI, maker of ChatGPT, announced the result
If that following some bullshit about how stupid humans are didn't tip you off, this should also make it extremely clear:
OpenAI privately contacted Litt, Sawin, Gowers and a number of other mathematicians to verify the LLM’s proof. Together (and without the company’s direct involvement), they wrote up their individual takeaways. (No external experts have seen the AI’s original output, however—just an edited version of its train of thought.)
This is a marketing campaign
This format is extremely common for trying to market LLMs as useful. These are all over the blog/news section of LLM companies. There's heaps of these in 'reputable' pop science and tech magazines and they're often summaries of internal 'research' the LLM company has done with an interesting lack of information; there's no output or history of the interaction with the LLM, and the standard of proof is trust me and the math nerds we bought, bro.
If you cannot read this article and understand it is a marketing campaign and that there is a lot of distracting fluff and exactly zero evidence to support the positive claims about LLMs, prestigious universities across the world offer free 50 level/introductory courses on English. Community colleges can help hook you up, too.
Functional literacy is a skill we can all improve.
Further entry level courses about how science works will also help you understand why it is actually despicable that this article was ever published outside the LLM company's blog and disguised as scientific journalism.
