[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] [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

Name
Options
Comment
Verification
4chan Pass users can bypass this verification. [Learn More] [Login]
File
  • Please read the Rules and FAQ before posting.
  • You may highlight syntax and preserve whitespace by using [code] tags.

08/21/20New boards added: /vrpg/, /vmg/, /vst/ and /vm/
05/04/17New trial board added: /bant/ - International/Random
10/04/16New board for 4chan Pass users: /vip/ - Very Important Posts
[Hide] [Show All]


[Advertise on 4chan]


Is the plateu a reasonable projection for AI? Even if it is getting 100x faster at making computer chips, that make it run 1000x faster, will it have any practical applications beyond being able to spit out a glue pizza recipe faster/using less energy? Ai doesn't have the sentient capacity for invention of actually new things. Physics seems keen on denying all the real world cool sci fi concepts like relativistic travel, intertidal/gravity dampers and laser swords.
>>
It does not understand concepts, it never will, it won't get any better than it is now, this tech has already plateaud
>>
You can't safetyslop a true AI.

Some autist will create AI God at home, not the industry.
>>
>>107013662
Perhaps, but will we ever get a true ai? Will the scope of our goals simply peter out to solving real world problems before that point?
>>
>there is no practical difference between an AI that can copy 99.9999% of Reddit posts and one that copy 99.99999999 of Reddit posts
Fuck I hate AI cultists so much. AI can be useful in so much as it's database contains useful references. No LLM has ever had a novel thought, or devised a new solution to a problem other than maybe trial and error.
>>
>>107013885
So, a search engine on crack that can do 1+1=2, so to speak?
>>
>>107013885
>No LLM has ever had a novel thought
it doesn't have thoughts, period. all they do is knead strings of text that maybe had thoughts behind them.
>>107013933
1+1 appears in training text a lot, so it's unsurprising that it will try to guess =2. more recently there are frontends that detect that there are numbers in the prompt and reroute the question to be translated to a scripting language, evaluated, and result translated back to human language. essentially a very inefficient way to open a calculator. it's all smoke and mirrors.
>>
>>107013933
>search engine on crack
>linter on crack
>quick back-of-the-envelope estimates
>quick sanity checks
>context-aware suggestions
it was never going to be AGI for a fundamental limitation of an LLM is their lack of world model
but people pretending it's useless also seem to be those complaining about what it can do
>>
>>107013585
Stop using the word AI and just say what it actually is, an LLM, and yes, LLMs are a dead end.
>>
>if you enlarge the engine your car will fly like a jet
Nope.
>>
>>107013885
>AI can be useful in so much as it's database contains useful references.
So many people don't get this. LLMs aren't even about language, it's a data optimization function. It's just automating the process of looking at data that already exists.



[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.