Stevvo

9 ms·

1.
by Stevvo 3d ago
That is not the narrative given in the OpenAI announcement. They claim to have started working on it after "we heard rumors that two Millennium Prize problems had been resolved", on September 1st, that "we later realized was
2.
by Stevvo 4d ago
They said they started working on millennium problems after hearing a rumor someone had found a solution to one of them. That's just bad taste. It indicates negative motivation rather than positive motivation from the start.
3.
by Stevvo 4d ago
No, it is not. Proving things without comprehending them built much of human civilization. Comprehension is a relatively new fad that came along in the 17th century with the scientific method.
4.
by Stevvo 6d ago
I don't get it. Surely if you were developing novel biological weapons, you would not use a hosted AI service where Anthropic can read what you are doing. And, why would you need to? Any chemistry graduate could make you dozens of high
5.
by Stevvo 7d ago
A lot of the time people were sharing transcodes because they didn't know any better. Same thing happens for video content today, e.g. most video on X looks like 240p because of it.
6.
by Stevvo 7d ago
I find that story hard to believe. Your industrial design class designed folding iPhones 2 years before anyone had even seen the first iPhone? Sounds like a leap. That said, at the time, the most popular phone in the world was a folding pho
7.
by Stevvo 8d ago
Don't be put off by the box asking for your "affiliation". I wrote "None", clicked submit and it took me straight to the Atlas.
8.
by Stevvo 9d ago
My high-school economics teacher taught us Georgism is the optimal form of taxation. He gave us a bunch of dense math to back up that claim, but I can remember none of it.
9.
by Stevvo 12d ago
I don't work at IBM but there are many things to admire. e.g. You can take software written 40 years and run on a new IBM system today without even recompiling it thanks to the Machine Interface. Have fun trying to run a 40 year old C+
10.
by Stevvo 12d ago
Interesting they are renting out agent skills. You can pick up "Java Modernization" at $20 a month for "Specialist workflows to help with modernizing Java applications", or "IBM i" for "Specialist modes, w
11.
by Stevvo 14d ago
This is completely useless for video game map blockout. You need collisions for level blockout, which this does not provide. It makes gaussian splats, not geometry.
12.
by Stevvo 16d ago
There is no puzzle in Portal 2 that requires fast reflexes.
13.
by Stevvo 17d ago
Fantastic as someone looking for an affordable vacation destination.
14.
by Stevvo 19d ago
Feel like Ferrari Luce was designed specifically to turn people away from the idea of electric super cars.
15.
by Stevvo 19d ago
The manufacturers certainly want the future of supercars to be combustion engines. Combustion engines are their moat; without them, there is little unique about their vehicles. But BMW felt the same about regular gasoline cars & hybrids
16.
by Stevvo 21d ago
Curious how different the safety outcomes will be to the US. A Waymo is safer than an American driver, but so is a European driver. A driver from Norway is even safer than a Waymo.
17.
by Stevvo 23d ago
Depends how you define fidelity. The mesh resolution in MSFS isn't any higher resolution. Looks about the same.
18.
by Stevvo 1mo ago
The benchmarking in the article gives a clear answer to the title question: Python and Javascript. However, the article doesn't just bury the lede, it misses it entirely, getting distracted by outlier results from clojure and j.
19.
DARPA Lift Challenge – Aug 7 – Main Broadcast [video](youtube.com)
2 points by Stevvo 1mo ago|0 comments
20.
by Stevvo 1mo ago
Silly anchoring is forbidden, could be a great sailing destination.
21.
Security Determination on the Threat Posed by Foreign-Produced Power Inverters [pdf](fcc.gov)
1 points by Stevvo 2mo ago|0 comments
22.
by Stevvo 2mo ago
Interesting they are still using "Mythos Preview" instead of "Mythos 5"; I had read from others who had access to both that Mythos 5 is less capable.
23.
by Stevvo 2mo ago
That's an odd connection to make. The depiction of passengers in Elite Dangerous is limited to some numbers on a UI screen.
24.
by Stevvo 2mo ago
Hard to imagine it as a passenger aircraft any time soon, but if they manage to get the demonstrator built it looks like it could make a useful aerial refueling tanker.
25.
by Stevvo 2mo ago
If you stop and think about the problem it really is quite simple. Just need to build a graph of the game state and then run A* to get to the end.
26.
by Stevvo 2mo ago
Why? 30% is passing the first two problems only, which are really very simple.
27.
by Stevvo 2mo ago
Fiber optic drones were last year's weapon of choice, this year it is drones with autonomous terminal phase. This happened because the AI data center build-out increased demand for fiber optics, driving up the price to less affordable
28.
by Stevvo 2mo ago
It passed the first two puzzles, which are incredibly simple but the bench doesn't explain what the goal is. Any model with a knowledge cut-off after the introduction of ARC-AGI-3 could probably pass the first two puzzles just by knowi
29.
by Stevvo 2mo ago
30% on ARC-AGI-3 is the first two puzzles. It cost $20000 in tokens to do that. That is a terrible result that doesn't imply anything.
30.
by Stevvo 2mo ago
As far as I can tell, the whole point of the backrooms meme is pointlessly hostile and counterproductive petty arguments about ideas of what "backrooms" means and who owns those ideas.

More ›