Do you have a source, especially of what was in the training data specifically?
As I mentioned, they would have to find out whether the mentioning of the approach is only present in Tristan Buckmaster's chats and not in other users', including free users', chats.
> no right to a trial before your rights of travel being revoked
I'm not sure what exactly this refers to (because of the paywall), but revoking right of travel before trial is fairly common, it the US it would be classified as detained without bail. And happens when there is a reasonable concern for flight risk before trial as seen by the judge.
Ngl the weird UI bugs made me think it must be made by AI, either that or the developer has very skewed frontend skills where gradients are fancy but sliders & interactions are broken.
I am actually going to guess that this is in fact unfortunately not an LLM. I have been made a few audio webpages and don’t think a lot of these mistakes would have been made, even though the model can’t hear :^)
<!-- EXACT setup from working simple-test.html -->
<!-- Chiptune3 now loaded via Vite build system -->
are pretty stark llmisms. also the random emojis in front of the duplicate and triplicate "Browse Music" links, the corny tagline in the footer, yeah nah. if this was made by human hands, it could only have been as elaborate satire. bad/inexperienced human code does not look like this.
> <!-- EXACT setup from working simple-test.html -->
All LLM-kind would be vastly improved if the words "exact" and "brilliant" were nerfed to hell in their pre-training weights or even just removed from their training distributions entirely. Virtually nothing outside of mathematics is "exact", and virtually nothing outside of colors should be described as "brilliant".
Also basic functionality not working paired with awkward and vaguely nonsensical phrasing - search, show all, "Showing tracks from all groups.", queue all -> "No songs found in this group." And that paired with some otherwise relatively sophisticated techniques involved in the basic functionality.
LLMs are on occasion like a kid that can do a standing double backward somersault, but can't walk. It's the same tell you'll have in discussions online. Somebody will spew out extremely esoteric (and accurate) knowledge/concepts, but then demonstrate a complete misunderstanding (or lack of knowledge) of basic concepts in the exact same domain.
nah, that's an entire 10 episode rewatch if this libertine vibe coder's agentic engineering pipeline holds! ... shit, mostly ads, if only the dude went through a few more testing cycles ... nah, that's enough time for some auto-research into the psycho-linguistic consequences of getting The Trump voted for president after the whitest black presidency in the history of the world, forever!
Infineon as well, but all those EU semi companies mostly make cheap low margin chips that more or less compete with the commodity chips of China and Taiwan but not with the IP from the US or even Israel who designs and sometimes also makes mostly high margin CPUs, GPUs, FPGAs, switching fabric, cellular modems, etc stuff that the EU uses and imports a lot but has no domestic players in.
In that regard yes, the EU semi industry is kind of a wasteland, considering how big the EU and how small its semi industry is.
Because if you go down the callstack eventually you won't get the await keyword anymore; you'll get the actual 'waiters' and 'wakers' which define your scheduling
military/industrial customers want this for tactical operations
fully discreet low latency compute that integrates with there existing space hardware
so the ability to run drones, and meatbots from space, with low hardware requirements on the ground, but with lets say 1000~2000 users max, power users, but not that many.
so batteries, heat sinks,he refrigeration radiators, on a sattelite fleet.
they have been trialing this for years and are pitching putting ALL of the military compute in space, no more dirty civilian hands on there toys
Sure, in the same way that there is unlimited solar power on earth.
There are orbits with 24/7 or near-24/7 sunlight - but those are very undesirable if you want a low-latency data link back to Earth. Just like you can get 24/7 sunlight on the North / South Pole - but they are still pretty bad locations for a data center.
LEO orbits like those used by Starlink have far better connectivity, but about the same sunlight exposure as the surface as the planet will be between you and the sun about half the time.
Also, power is the easy part. Cooling is far harder.
They did check and it was in the trainingset but they could not confirm if it was the /cause/ of it, which is a bit of a cop out.
reply