It’s kind of crazy really. Before LLMs, any type of world scale disaster you’d hope for what? Wikipedia backups? Now, a single LLM ran locally would be much more effective. Imagine the local models in 5 years!
There's a lot more than just Wikipedia that gets archived, and yes, that is a far more sensible way to go about it. For one thing, the compute required to then read it back is orders of magnitude less (a 15 year old smartphone can handle it just fine). For another, you don't have to wonder how much of what you got back is hallucinated - data is either there or it's corrupted and unreadable.
The processing required to run current language models with a useful amount of knowledge encoded in them is way more than I imagine would be available in a "world scale disaster".
What was interesting to me was how expensive this thing was, I wonder how much of that cost came from the software. Elon mentioned how they're trying to create a "world model" for their Tesla FSD, which has enabled them to use the same software for Optimus.
I'd imagine the future will be large "ready to go" Ai packages, which companies would tweak to their liking.
Seems to me that smaller, more focused models would be more practical. Why would the tulip machine need to know the rules of canasta or the history of peanut butter and jelly sandwiches? I wonder if there's going to be a persistent monolith/microservices cycle.
Wasn't it 180.000 EUR? Once you start buying equipment for the industry you will discover that consumer price levels do not apply here. Perhaps it is the lower scale, or it is the burden of compliance.
This is such an interesting use of semantic representation. I wonder if it could be used to map out cases vs outcomes, and determine sentencing outliers.
You can still be sued. At that point, you either:
a) defend yourself as normal, while attempting to maintain anonymity (difficult & doesn't solve anything)
b) let the case go to default judgement (for likely an enormous amount of money) at which point Nintendo has every reason to try to track you down in real life, and can use the courts to help them do so. Hope you were really, really thorough about your anonymity.
File a lawsuit against a John/Jane Doe defendant(s) and describe who the defendants are, e.g., the people who code and maintain the emulator. Maybe a little harder than if you had the names but not by a lot.
I should say: the real difficulty is service, but if there's any way to get in contact with you, there's at least a chance it will be accepted as good service. There was famously a recent case where a defendant known only by a Bitcoin address was served by sending a transaction with an attached message to that address.
It can be a resume piece. Having been in these communities (server emulation instead of console) you find some insanely bright developers in them. A lot of them don't have a formal education background, but are willing to build out well thought out pieces of software. You just have to do the research on them if they tell you this is who they are, a lot of the times you can find out plenty about how they code and design things out in the open. I'd probably almost always hire such a dev if their code seemed good, and I knew they cared about the quality of their code. Having been involved in those communities though I do know there are awful devs who don't care about what anyone says "my code aint broken" types who never give credit when you help them either (reputation / respect matters more when you're coding for free in some of these communities).
I'll double down on personality that parent comment pointed out. If you see someone submit a project they worked on instead of formal education then you need to see what their exact contribution is and how they interact with other devs. Look at past github comments, issues, etc. Lurk in their chat server or pull logs. Expect to see a more casual attitude than what will show up at work, but pay attention to how contributions are handled, how credit is handled, and how disagreements are handled. It's not a universal problem, but I've seen some of the worst developer drama around volunteer open source video game projects.
Do they expect a company to believe them when they say "Yes, that anonymous social profile is actually me, John Emu maker. Really." Then if they proove it, by logging in and posting a known sentinel for example, do they expect everyone on that interview loop to keep their secret?
Companies generally don't take anonymous credentials.
This is a little too risky since it could wind up with someone revealing their identity inadvertently or otherwise. If you truly want to be anonymous you should not be leaving so many breadcrumbs. All it takes is for a former employer to have a security breach and someone to point out that they found your real identity.
Because making an emulator takes a lot of work. And emulator devs want some compensation for this work - be it money (in form of donations) or a status that could help them gather followers or even land a job.
Nothing you wrote has a spec of evidence. The current administration has been clearly targeting Elon Musk and Tesla/SpaceX, so why on earth wouldn't Elon support their opponents?
everyone with a logical mind would see it. your political opinion should not cloud your judgement… people like you are the reason we have a senile old man leading our world to wars. I cant fathom why americans hate trump so much but are so lenient of biden and democrats that sent us to the worst conflict the world has seen in decades
