[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/biz/ - Business & Finance

View post   
File: 170 KB, 360x346, a20ddd95ed33c6937246c1f37232e519.gif [View same] [iqdb] [saucenao] [google]
57687863 No.57687863 [Reply] [Original]

>People actually thought AI was just a bubble

>> No.57687898

>>57687863
Can someone explain to me where the concept that artificial intelligence was a bubble came from? All the specialized media came out to clarify that what we were seeing was only the surface of the level of development that was possible, that was two years ago and since then it is not like there was a stop of advances in fact it was the opposite, we had more and more things related to the ia (generative) taking a predominant role in our day to day life.
And now that two years have passed and projects that needed extra time are finally becoming commercially available (openia and its video generator) or passing the patent stage for release (tau and its artificial intelligence logic) or simply reaching a higher level of development (NVIDIA) everybody is surprised because they thought it was a bubble?

>> No.57687976

>>57687863
I work at a company deep in the ml / big data / data science. world. We are very successful. Our next big product is a fucking chatbot. Yeah this shit is not a bubble

>> No.57687978

>>57687863
It is - AI will be widely used, but the market is acting like we're going to have robot maids and sexbots within 5 years.

>> No.57687993

>>57687978
Nope, instead all those little things that gave one meaning will be lost to automation and efficiency until there is nothing but a detached, depersonalized, derealized husk of flesh and bones in this filthy jewish world

>> No.57687998

nvidia will go 1.5k by the end of this month, mark my words
ai is the new industrial evolution

>> No.57688022

>>57687863
When I would post about AI here years ago, glowies and dummies would call it vaporware and say we had nothing to worry about

>> No.57688029

>>57687976
>millionth chatbot
so, it's a bubble than

>> No.57688043

>>57687863
It kind of is for now. What exactly can I do with AI that is practical right now?

>> No.57688049

>>57688043
The main application of it is propaganda purposes if you ask me.

>> No.57688051

>>57688029
1+1=6

>> No.57688065

>>57688043
Nothing really yet, the promise of AI is that it's going to put everybody out of a job which will force the government to institute a universal basic income that will immediately go straight to corporations

>> No.57688069

>>57688049
Using AI for propaganda or AI is propaganda?

>> No.57688077

>>57687978
Heh theyre wrong as usual.
It's 3 years

>> No.57688084

>>57688043
Nothing, the general public will never have access to unerfed AI. Even Tesla had to make adjustments to their self-driving cars because of liability, self-driving without shit like putting your hands on the steering wheel was already achieved years ago but the company can be in deep waters if people started dying while sleeping inside their auto drive cars.

>> No.57688087

>>57687898
>tau and its artificial intelligence logic
You are seriously retarded

>> No.57688116

Chatbots do a lot; there are already agents that can do all kinds of web stuff for you. Just a matter of time desu
Lots of startups making AI employees already

Holy fuck the pause & captchas to post is so shit

>> No.57688134

>>57687976
Same here. We had a meeting recently for ideas related to LLMs and half of the company showed up with their gimmicks.

>> No.57688136 [DELETED] 

>>57688116
Chatbot isn't really AI, it's just predefined script, facebook, discord has chatbots for years. If you're talking about LLMs only huge would be able to do ot because of resource drain even then right now it's only OpenAI that has the best one.

>> No.57688145

>>57687863
It's only the begining, get on GRT before it's too late.

>> No.57688153

>>57688116
Chatbot isn't really AI, it's just predefined script, facebook, discord has had chatbots for years. If you're talking about LLMs only huge companies would be able to do it because of resource drain even then right now it's only OpenAI that has the best one.

>> No.57688156

>>57687863
"was"?

>> No.57688220

>>57688069
The former.

>> No.57688298
File: 55 KB, 800x600, AGIX.jpg [View same] [iqdb] [saucenao] [google]
57688298

>>57687863
>>People actually thought AI was just a bubble

It is going to revolutionize every part of society and the economy over the next decade. Losers only think of what it cannot do now. Winners can extrapolate from current capabilities.

>> No.57688318

>>57687898
The bubble is based on speculation that this is just the beginning. That chatbots and image generators are just the early incarnations of an AI revolution that will upend civilization and usher in a new epoch for humanity. The truth is that what we have now is pretty much all we're going to get out of transformers. There will be minor improvements on existing chatbots and image generators, but there's nothing in the pipeline that will give us the next great leap forward in 5 years like the transformer was in recent years. We will soon retreat into another AI winter before we emerge in 10 years with the next great innovation.

>> No.57688320

>>57688153
I'm not talking about the old-timey chatbots
I mean there are chatbots that can get company specific info or scrape info about you and write a unique message to each user.
Just look at all the AI SDRs or customer support agents coming up; you don't need a terribly fine tuned model or anything either

>> No.57688322

>>57687863
Just a bubble? Lol , you said it now and you are going to regret if soon, mfer. Im sure AI and RWA is the future of crypto. I always do my research for an AI coins and take a look more of rwa in eesee. I dont know if its just me, but i believe it.

>> No.57688391

>>57687863
AI as you know it is a bubble and functions as a tool to prop up stock valuations to defend against USD collapse. the problem is energy. both the dwindling recoverably cheap energy underwriting productive growth and energy efficiency of computing systems that, while massively parallel, fail to approach the efficiency of biology. until that time, it will always be more efficient to hire a human to do real cognitive tasks.