western regimes trying to put up a facade of progress that really changes very little

Yes I agree. I guess this is at least trending towards positive as they have to pretend to do something, things are really fermenting in the core. But whenever I hear western politicians talk about taking a “hard stance” towards Israel it usually boils down to sanctioning some random government officials (matters little) and at most “voting for sanctions on the illegal settlements”. There is never a good answer to the follow up question “Since the Israeli government if facilitating and de facto implementing these settlements, shouldn’t they also be sanctioned?”. It is so pathetic I can puke.

I think the drama is a bit two-fold. There is the question of whether OpenAI scraped/used the human researchers chats to gobble up a result in just two days (one of the researchers spoke with OpenAI two days prior and then suddenly they present a proof using the same stuff).

Then there is a whole other drama of OpenAI trying to pressure Buckmaster to dissociate with Alpöge because he works for Anthropic instead of OpenAI etc etc. Which obviously is frowned upon.

There is probably more but that seems to be the gist of it from what I can tell. Probably will be developments throughout the day.

I am on the other side of analysis, so def not an expert on NS, but Alpöge and Buckmasters work seem substantial, and as with all mathematics it heavily builds upon previous ground work of Diego Córdoba and Martínez-Zoroa. MZ particularly has made substantial progress in how we can construct forced blow ups.

In theory there might not be anything preventing their methods for working on the full NS, but it is thought to be insanely technical. Possibly something that LLMs and throwing A LOT of compute at could solve.

OpenAI is doing big-tech things here. Please also read https://cims.nyu.edu/~tristanb/statement.pdf and https://mathstodon.xyz/@tao/117233528517340774. A lot of people in the field has done similar things to this, the interesting result is not necessarily coming from OpenAI in this case.

Generally, these companies are not only throwing billions of dollars worth of compute at these problems, they are also employing top-tier HUMAN mathematicians (some of the worlds best) to work on these problems using their internal tools. Then when things are getting published, the human part of the equation is usually reported on…

From what I understand it could be nothing or very big. Ban only primary trade from illegal settlements = sounds like something, but is a very little part of total trade with Israel. Ban primary + secondary = ban of 99% of all trade since almost all Israeli products have some component or likewise originating from industry in the settlements.

Huh, nothing new then?

Oh that’s easy, don’t let the fascist government host or point nukes at China then. Very low risk of that accidentally happening, don’t you think?

Yea I actually did read some of his articles and shorter works, but found them quite heavy. I hope his books are a bit easier to follow.

Uhh, I am so behind on my book list. At the bottom there is just a line that says “Every book by Losurdo” lol, I guess I’ll get to it