pull down to refresh

Anyone else get really bummed during a bull run? (An update from frostdragon)
1496 sats \ 17 comments \ @frostdragon 5 Mar bitcoin
https://m.stacker.news/18991
Been a while since I've posted around here. I absolutely love the excitement and the energy in the air, and at the same time I'm thinking "please don't hit ATH, please don't hit ATH, no god please NOOOOOOOOOOOO". So conflicting. I'm not one for bitcoin price predictions, but I think <$30k days might be behind us.
Anyway, how's everyone doing around here. Just left unchained a few days ago since they're trying to charge me freakin $250/yr to use multisig. I used to recommend them a lot - I really think the collaborative custody model is one of the best ways to go for long term holding, but here I am with a self run multisig setup now. Somewhere my nemesis @DarthCoin is laughing maniacally. (I still think hiding a seed phrase stenographically in an image on the open internet is a terrible way to go, buddy).
Do people still use Nostr/Damus? That was fun for a while, but I'm not so convinced the incentive for people to run their own servers is really there for long term scalability. Maybe it is, and if my loss of interest is causing me to miss out, someone feel free to correct my ways.
Anyone else around here as anti AI hype as me? I might be missing something, but I'm just not seeing AGI as a real possibility just because ChatGPT can regurgitate google results. Don't get me wrong, it's a game changer in so many ways, but the more I try to use it, the more I see its limits. It's can't problem solve. It can't come up with new thoughts. It can't make something that hasn't been made before. It's very limited in its coding abilities (yeah yeah it can make a basic website with some basic features that exist literally everywhere, and hey, that's huge in it's own way, but try to come up with a brand new kind of functionality and get it to build that. In my experience, you get a Frankenstein made of bandaids on bandaids on bandaids, which is a perfect recipe for security vulnerabilities). But maybe I just haven't given it enough of a chance.
That's all for now. Maybe I'll stick around here for a few weeks before going dark again.
write
preview
related posts
501 sats \ 6 replies \ @south_korea_ln 5 Mar
Anyone else around here as anti AI hype as me?
You and your nemesis might have more in common than you think.
Jokes aside, I'm also thinking it's very much overhyped, at least from a stock and investment perspective. I do use it daily, but for mundane tasks that would take me longer to do myself. Debugging too.
It still has ways to go before being useful for the truly creative tasks.
reply
51 sats \ 5 replies \ @joda 5 Mar
Agreed. I think we were all shocked by the visual "creativity", but many of us are disappointed in the lack of creative ideas.
Like try asking it for a bitcoin scaling solution that doesn't compromise the trilemma.
reply
21 sats \ 4 replies \ @frostdragon OP 5 Mar
I would even say those that are shocked by DALL-E haven't spent a lot of time trying to get it to do a very specific thing.
Like, try to imagine a fictional character or creature and make it as detailed and specific as possible (pose, clothes, accessories, features, viewer perspective, etc), and try to get AI to generate it exactly how you envision it, it turns out to be way more difficult than you might think. Maybe I'll post a bounty with some specifications for an image idea just to try to prove that point.
reply
50 sats \ 1 reply \ @Atreus 5 Mar
I gave up on it. It's good at getting “an image,” but not the image I have in my head. Unfortunately, it's become harder to source man-made images since DeviantArt, ArtStation, et al., are overrun by AI generations.
I've been looking for a good place to source human-made scifi/cyberpunk images for my newsletter and it's annoyingly difficult.
reply
0 sats \ 0 replies \ @frostdragon OP 5 Mar
It's good at getting “an image,” but not the image I have in my head.
exactly
reply
100 sats \ 0 replies \ @frostdragon OP 5 Mar
Here ya go: #451746
reply
0 sats \ 0 replies \ @south_korea_ln 5 Mar
Fully agree. I was confronted with its limitations when I was following my son's instructions to make the robot he had in mind. As he does not understand the underlying tech, he was the most critical customer one could imagine.
E.g. he was asking the robot to have square eyes (as that's what he saw in a robot from the toystore), yet the generated images kept showing round eyes. Even though ChatGPT explicitly told me "I took special care of including square eyes as you asked". And if I ever managed to include something my son asked to change, it changed many other things as well. Understandably, as that's how it works, but a hard sell at this point to convince this tech can fully replace humans.
My son will not be returning customer for now :)
reply
150 sats \ 2 replies \ @OneOneSeven 5 Mar
The puzzle solver returns lol, been a while.
I tested three LLMs the other day with a few sentences with all lowercase words and capital letters strewn about and asked if it could spell out what the capital letters said. Gemini, ChadGPT and Copilot, they all failed spectacularly, I even tried helping but it was like talking to a shitcoiner.
Meanwhile Elon suing OpenAI saying they have AGI capabilities.
Just give us the AI to make droids to help (and protect) humans in the physical world.
reply
63 sats \ 1 reply \ @frostdragon OP 5 Mar
Anytime I hear anyone say anything about AGI
https://m.stacker.news/19020
Like, there's no freakin way we're anywhere close to AGI, right? I'm very open to being corrected here, but like, at the moment I'm not seeing it.
@OneOneSeven, thanks for the welcome back
reply
42 sats \ 0 replies \ @OneOneSeven 5 Mar
Good point, I had to look up what it's supposed to mean from the guy who coined it: https://www.youtube.com/watch?v=kMUdrUP-QCs
The definition I use now is that [AGI] is a system that can do all the cognitive kinds of tasks that people can do, and possibly more"
Pretty vague and general, but definitely far from where the current LLMs are now, per my example. But who knows what goes on behind closed doors at ClosedAI.
reply
50 sats \ 0 replies \ @Atreus 5 Mar
After deep diving the technology and really observing how it gets made—lots and lots and lots of human intervention needed—I consider "AGI" just a technocratic superstition, and probably a psyop to scare the shit out of everyday people.
This civilization collects mountains of data, though, and AI is very good at querying large databases to inferring mathematically probable answers to a question, such as "What's the next most likely word in this sentence" or "Which of these citizens fits a criminal profile?"
The implications of a technocratic society with a centralized AI feeding off gobs of data produced by the plebs are unpleasant, to say the least.
reply
20 sats \ 0 replies \ @Voldemort 5 Mar
Anyone else around here as anti AI hype as me?
I think it is overhyped in its current state, but it will provide a lot of value over time. Think of the internet, it wasn't ubiquitous until we had things like internet explorer and then google and whatnot.
I think AI will be similar but to a smaller degree. At some point something will be built on top that increases its utility.
reply
0 sats \ 0 replies \ @riberet19 6 Mar
I get discouraged because I know that all the nocoiners come to ask me nonsense and ask me how much money I have made or what they want to invest in Bitcoin, it is an endless cycle, but hey, what are we going to do...
reply
0 sats \ 2 replies \ @ZezzebbulTheMysterious 6 Mar
Re AI, I am most impressed with Image and soon to be video generation.
Stable Diffusion XL and various LoRAs have brought me many hours of joy.
You can get started on your own hardware with this: https://github.com/AUTOMATIC1111/stable-diffusion-webui
reply
0 sats \ 1 reply \ @frostdragon OP 6 Mar
Curious to see if you can beat my challenge #451746
reply
0 sats \ 0 replies \ @ZezzebbulTheMysterious 6 Mar
WDYT #452136
reply
0 sats \ 0 replies \ @SaToSHiWoMaN 6 Mar outlawed
stackers have outlawed this. turn on wild west mode in your /settings to see outlawed content.