[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vm / vmg / vr / vrpg / vst / w / wg] [i / ic] [r9k / s4s / vip / qa] [cm / hm / lgbt / y] [3 / aco / adv / an / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / pw / qst / sci / soc / sp / tg / toy / trv / tv / vp / vt / wsg / wsr / x / xs] [Settings] [Search] [Mobile] [Home]
Board
Settings Mobile Home
/g/ - Technology


Thread archived.
You cannot reply anymore.


[Advertise on 4chan]


So called "AI" is a meme at this point. It's just really good predictive text. Don't get me wrong the technology is impressive but ultimately its just good predictive text (plus the summary, rewriting & image generation stuff. All of that is so, so far away from AGI and all the AI companies nowadays are laser focused on narrow expert systems. Oh and for the love of god stop calling it AI, it's just machine learning, sure some of the stuff with the newer models is impressive but it's still really narrow. You can't apply ChatGPT to the task of learning how to do maths, or how to play chess. i.e. it can't learn anything from first principles. There are uses for it that can augment parts of your workflow but nothing like people say and nowhere near as much as people thought. Hence why investment into AI dropped last quarter after people realised its not magic beans.
>>
what a load of AI generated words, too bad I don't have an AI to read them
>>
So AI is just "really good predictive text"? No shit, dude. Let's downplay one of the most game-changing techs ever and call it fancy autocomplete. Yeah, AGI isn’t here yet, no fucking duh, but acting like the progress in narrow AI isn’t a big deal is like looking at the Wright brothers and saying, "lol, can't even fly to the moon, what a joke." It's called progress, not magic beans.
Investment dipped? Boo-fucking-hoo, maybe because the Wall Street suits realized they can't milk their golden goose overnight. And please, enough with the "it's just machine learning" bullshit. That's like arguing over what color to paint the rocket while others are already strapping in.
So keep shitposting and watch the world leave your sorry ass behind while you're busy nitpicking over what to call the damn thing.
>>
>>101953307
>Not there yet.
Not gonna get there anytime soon either mate. This sort of stuff is way more interesting and coincidentally would be a way to increase the speed at which machine learning advances:
>https://en.wikipedia.org/wiki/Spiking_neural_network
>https://en.wikipedia.org/wiki/Wetware_computer
>https://techcrunch.com/2024/04/15/investors-are-growing-increasingly-wary-of-ai/
>>
>just really good predictive text
Actually better than a lot of regular people. Tell some rando to give you a five paragraph story about Hitler as the Easter Bunny delivering ice cream cones to Aum Shinrikyo and most of them will seize up. The few who do will 99% of the time produce something inferior to the LLM text. That it's not Cormac McCarthy is irrelevant, it's still doing better than most humans are capable of.
>>
>>101953267
No you see, I actuallty wrote that myself because I'm not retarded.



[Advertise on 4chan]

Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.