daveguy

22 ms·

481.
by daveguy 11mo ago
I can't wait for the first company to do this. I hope someone comes around to do a good post mortem from the rubble.
482.
by daveguy 11mo ago
> ...I can't imagine an algorithm suggesting that content. It seems like a default suggestion. This implies that the default suggestion isn't a data analyzed soup of what people of a given age / location / demographic
483.
by daveguy 1y ago
I'm not going to the mat for anyone. I'm just saying AI use in spreadsheets is a terrible idea because AI just isn't that good. Currently I'd put it worse than tearing things up for ${hotLanguage} because at least ${hotL
484.
by daveguy 1y ago
It may not be what you meant to say, but it's exactly what you are saying where ${hotLanguage} is the latest automation platform or AI gimmick.
485.
by daveguy 1y ago
It's not what you had in mind, but that's what you get. Because automation, integration, and AI are currently garbage -- Salesforce, Netsuite, doesn't matter. They don't do the magic that they promise. Because process is
486.
by daveguy 1y ago
Musk definitely did several nazi salutes. Even did it with flourish like the hitler videos. Pretending otherwise makes you look like a sympathizer or a fool. Sorry.
487.
by daveguy 1y ago
Is that eating fish that fish eat or eating fish that eat other fish? And why?
488.
by daveguy 1y ago
It was exfiltration -- copying or moving data from an internal system to an external system. They insisted on and bragged about full access because now it would be "efficient". But it was clearly just simple opportunity for theft
489.
by daveguy 1y ago
Welcome to the other side of your propaganda bubble.
490.
by daveguy 1y ago
They are both designed, trained, and evaluated by how well they can predict the next token. It's literally what they do. "Reasoning" models just buildup additional context of next token predictions and RL is used to bias outp
491.
by daveguy 1y ago
Most everyone. Rural Mongolians arent getting their news from US propaganda. Oh, and how could I forget the most racist propaganda spewing one of all, X. People who get info from these sources have been manipulated to shut down their own cr
492.
by daveguy 1y ago
Much less well understood by folks who get their news from Fox News, News Max, or bro podcasts.
493.
by daveguy 1y ago
*as long as there's no tariffs involved and no entrenched power monopolies taxing folks for installing solar. Looking at you, Southern Power Company.
494.
by daveguy 1y ago
I wouldn't bother. I've got no use for crypto.
495.
by daveguy 1y ago
Literally anyone can make up a Blockchain. That's why there's a convoluted mess of crypto.
496.
by daveguy 1y ago
(and on to real DCs later) would imply their bare metal is now located in a data center.
497.
by daveguy 1y ago
Except any country can make up their own crypto "stable" coins.
498.
by daveguy 1y ago
Confidence in crypto will go poof when there's a big enough run that the measly 7 transactions per second can't keep up with demand. We just haven't seen a crypto bank run yet. The $100 / transaction is nothing compared
499.
by daveguy 1y ago
> We've been fighting for 30 years about MMR and autism, which was a single retracted study discredited by every study since. And how we're gonna somehow discuss in courts about the correct statistical model to explain a real c
500.
by daveguy 1y ago
Just replaced a Samsung fridge. It was the worst I have ever had, and it wasn't even kludged up with smart-ai-internet-advertisement bullshit yet. The compressor went out after about 12 years (which is apparently good for current refri
501.
by daveguy 1y ago
You apparently do not understand the difference between causation and correlation. This is also why the execs were concerned -- because most Americans are the same. 1. Diagnosis rates. Because diagnosis criteria has changed. 2. Sure okay. D
502.
by daveguy 1y ago
Your first sentence says "there is a credible link between them" and the rest of your post points out how the link is not scientifically credible. Were the first and third paragraphs just bait? You can't claim that "scie
503.
by daveguy 1y ago
> I think with ARC-AGI 2 this was the kind of compute budget they spent to get their models to perform on a human-ish level. It was ARC-AGI-1 that they used extreme computing budgets to get to human-ish level performance. With ARC-AGI-2
504.
by daveguy 1y ago
> This is only helpful when there is a professional therapist available soon enough and at a price that the person can pay. In my experience, this is frequently not the case. That doesn't make a sycophant bot the better alternative.
505.
by daveguy 1y ago
It parses however the Trump administration wants it to parse in any particular context on any particular day. Their legal moves have been a shit-show of incompetence and callous disregard for the law.
506.
by daveguy 1y ago
*Roughly 40% of the country, give or take. Don't be complacent.
507.
by daveguy 1y ago
> Spreadsheets are already a disaster. Yeah, that's what OP said. Now add a bunch of random hallucinations hidden inside formulas inside cells. If they really have a good spreadsheet solution they've either fixed the spreadshee
508.
by daveguy 1y ago
Agreed. Currently is the key there. I was referring to whent a run starts. It's been at $100 before and we haven't even seen a serious run yet. $100/tx is the "every few years" rate.
509.
by daveguy 1y ago
The Robinhood one is a direct copy of one of the original movie posters. The others are a lot more interesting, imo.
510.
by daveguy 1y ago
The implication wasn't to use the raspberry pi toolchain. Just that toolchains are required and are a critical part of developing for new hardware. The Intel/AMD toolchain they will be competing with is even more mature than rpi.

More ›