machdiamonds

5 ms·

1.
by machdiamonds 2y ago
I don't use my Quest 2 because it's so annoying trying to keep the screen in focus with the fresnel lenses.
2.
by machdiamonds 2y ago
Pretty funny to read the comments from xAI's initial announcement now. https://news.ycombinator.com/item?id=36696473
3.
by machdiamonds 2y ago
Pretty simple explanations for all of those: - xAI opens sources models with a 6 month lag, look at Grok 1 - No one else stopped development, so why should he? - He owns Twitter, why wouldn't it be okay for him to train on Tweets?
4.
by machdiamonds 2y ago
Shouldn't be surprising with how hard everyone was pushing remote work.
5.
by machdiamonds 2y ago
He said they were using Bing and Google on the Dwarkesh podcast.
6.
by machdiamonds 2y ago
Do you actually have any examples of FSD v12 hitting emergency vehicles?
7.
by machdiamonds 2y ago
I think he said they would open source models after they have been released for 6 months or something like that.
8.
by machdiamonds 3y ago
It's not too hard to believe it is a coincidence when the most followed person on a platform shows up in your feed, especially if you follow tech accounts.
9.
by machdiamonds 3y ago
It's hard to guess these cards' real performance uplifts. According to Nvidia, H100 is 11x faster than A100, but that's definitely not true in most cases. If Gaudi3 is legitimately 4x faster than Gaudi2, it should be a very g
10.
by machdiamonds 3y ago
Hi, what differentiates double from Cursor?
11.
by machdiamonds 3y ago
Not sure why people on HN can't understand that companies actually need to make money to survive.
12.
Toyota recalls 1M US vehicles over sensor that could short circuit(reuters.com)
2 points by machdiamonds 3y ago|0 comments
13.
by machdiamonds 3y ago
It's definitely not by accident, it would be very easy to turn off community notes for tweets flagged as ads.
14.
by machdiamonds 3y ago
There are still some good subreddits, like "LocalLLama", but most of them seem to have been ruined by mods. I think this was the case even before the whole API issue, which just accelerated the decline.
15.
by machdiamonds 3y ago
You guys just believe whatever you read. Things that can easily be debunked with common sense. For example, there's a lot of yellow in this factory tour he did: https://www.youtube.com/watch?v=mr9kK0_7x08
16.
by machdiamonds 3y ago
Emad (Stability AI) thinks it's a 300B model https://twitter.com/EMostaque/status/1727373950685200674
17.
by machdiamonds 3y ago
"journalism"
18.
by machdiamonds 3y ago
Just like for humans, in terms of copyright, only the output should matter, not the training data.
19.
by machdiamonds 3y ago
It is delusional to think LLMs won't reduce the number of SWEs required. This should have been obvious from the first time you used GPT4.
20.
by machdiamonds 3y ago
There is no reason that AI can't be better than humans at requirements elicitation and communication. AI can have necessary domain knowledge about the company and could absorb extensive information provided. It could use analogies to e
21.
by machdiamonds 3y ago
Also, be born with an extremely IQ.
22.
by machdiamonds 3y ago
Actual translation: Tesla doesn't want people scalping their cars, and are using a technique many other manufacturers do.
23.
by machdiamonds 3y ago
Anthropic doesn't care about consumer products. Their CEO believes that the company with the best LLM by 2026 will be too far ahead for anyone else to catch up.
24.
by machdiamonds 3y ago
It's obvious Twitter was overstaffed but xAI has less than 20 employees so definitely not the case there.
25.
by machdiamonds 3y ago
Oh, the horror, people will be able to read things they can Google!
26.
by machdiamonds 3y ago
Leave it to HN to be negative about a team going from nothing to training a model competitive with a world class lab like Meta in 4 months.
27.
by machdiamonds 3y ago
fourth example: your comment
28.
by machdiamonds 3y ago
It's interesting how the bias once favored large companies that paid for advertisements. However, now that most news outlets are subscription-based and don't want to lose subscribers, they strongly bias their content to what their
29.
by machdiamonds 3y ago
I'm almost certain that the bandwidth cost is much higher than what they gain for recommendation algorithms. It's like saying "Yeah, that guy steals from your store, but at least now you know what's in demand!" (Not
30.
by machdiamonds 3y ago
Seems like a win for YouTube, they don't have to waste bandwidth on users that don't bring in any revenue.

More ›