Technology @lemmy.world Dadifer @lemmy.world 1y ago

OpenAI's reported 'superintelligence' breakthrough is so big it nearly destroyed the company, and ChatGPT

www.techradar.com OpenAI's reported 'superintelligence' breakthrough is so big it nearly destroyed the company, and ChatGPT

'Dr Frankenstein to the red courtesy phone please'

106

106 comments

Geez the reporting around this has been ridiculously sensationalist
- You mean OpenAI didn't just create a superintelligent artificial brain that will surpass all human ability and knowledge and make our species obsolete?
  
  The funny thing is, last year when ChatGPT was released, people freaked out about the same thing.
  
  Some of it was downright gleeful. Buncha people told me my job (I'm a software developer) was on the chopping block, because ChatGPT could do it all.
  
  Turns out, not so much.
  
  I swear, I think some people really want to see software developers lose their jobs, because they hate what they don't understand, and they don't understand what we do.
  
  You laugh now, but just you wait. If it turns out they've created a hyperintelligent waifu/husbando, this will inevitably lead to plummeting birth rates and the end of human civilisation.
- Superintelligent AI Just Pried the Keyboard from my Cold, Dead Hands
- Because that's the version that gets posted and gets clicked on.
  
  A dry technical writeup looking at the name of the project and how it indicates this is a different approach more in line with DeepMind's work and what that means in the context of doing high school level math is going to be interesting to only a handful of people.
  
  But an article that's contentious and gets hundreds of comments about how "AI is BS" to "AI is dangerous" all arguing with each other drives engagement.
just bs. They are trying to come up with an explanation for why altman was fired that is not: we caught him doing lots of illegal stuff.
- I think it's a hype move at this point. Like the guy who claimed he believed google's chat bot was sentient.
  
  I read another article that stated they had a computational breakthrough, in which their program can now carry out basic grade school math. No other model is able to actually carry out math equations, not even basic arithmetic.
  
  This is a significant development, but it's not like they're on the cusp of developing superintelligence now. I bet they are taking this small inch towards superintelligence, and hyping it like they've just huddled miles forward.
  
  The thing is, this could actually be a several miles jump. But where they want to go is not the grocery down the road. They are trying to fly to another galaxy. This is more like hyping up that you are going to land on the moon next year, at a time when you just figured out that rubbing two sticks together it makes a fire. Technically it's truly a leap, but we are so far away still.
  
  Is that the chatbot that they had to shutdown cause it wandered a little bit to much in 4chan?
  
  The worst part about is it that there have been already two winters in AI development, in the early twothousands and sometimes in the 70/80s? I think? because of exactly this: They always hyped up AI and said they'd solve all the world's problems in a short time, and when that obviously didn't happen, people got disappointed in it and pulled funding…
- I’m inclined to agree. It just stinks.
Connecting superintelligence to the board's recent actions, which Sutskever initially supported, might be a stretch.

Why do you do that in your headline then?
- Clickbait.
But can it open the pod bay doors?
- It can't do that, Dave.
  
  Dave's not here, man.
- Take your upvote and go watch more artsy 60's scifi, you brilliant sod.
  
  Is this a dave reference?
So was it all just a marketing stunt?
- CEO ousting shenanigans = 📉
  
  Release rumor = 📈
  
  They're not publicly traded, but I assume public sentiment still has an effect on things (ex. Partnerships, users buying memberships, etc.)
Hope it replaces the most expensive job position: CEO
What are these con artists up to? And why are so many people self replicating the propaganda?
- self replicating the propaganda?
  
  You can't self-replicate anything other than yourself. You replicate things; we use "self-replicating" because it's shorthand for "thing that replicates itself."
The "superintelligence" in question: the same old tech, but with a larger context window, which will make it hallucinate a bit less often.
- Not really.
  
  If it's name is Q* then it seems likely that it's a combination of Q learning and A* search, which indicates that this is an approach similar to DeepMind's AlphaZero as opposed to a transformer based LLM.
  
  In that context, getting it to be able to solve high school level math questions is pretty nuts.
  
  Though the details matter and right now all the articles discussing it are missing those, so we'll have to wait and see.
Alright, so the article really doesn't prove anything, just says OpenAI claims something and then fills it with words.

Let's be clear here; we don't even have an AGI. That is to say, artificial general intelligence, a man-made intelligence that is at least as capable and general purpose as Human intelligence.

That would be a intelligence that is self aware and can actually think and understand. Data from Star Trek would be an AGI.

THESE motherfuckers are now claiming they made a breakthrough on potentially creating an SI, a super intelligence. An artificial, man-made intelligence that not only has the self awareness and understanding of an AGI, but is vastly more intelligent than a Human, and likely has awareness that surpasses Human awareness.

I think not.
Why do I keep looking at these threads? The way people talk about this stuff on all sides is so asinine. Nearly every good point is accompanied by missing a big one or just ricocheting off the good one, flying off into space and hitting a fully automated luxury gay space commulist. Hopes, dreams, assumptions, and ignorance all just headbutting each other and getting nowhere.

Oh yeah, I wanted to know what "superintelligence" was and whether I should care. Welp.
- I think the takeaway is that they're trying to create a LLM that can answer questions that it wasn't trained on.
Almost sounds like the whole thing was a performance.
Yawn.

Let me know when we get a real Terminator or Matrix or Space Odyssey situation.
- heck, I'd settle for half a Short Circuit
- 🐈‍⬛
  
  🐈‍⬛
I'll believe it when Judgement Day happens and I die ~~in nuclear fire~~ when the wi-fi turns against me.
I for one welcome our new ~~robotic~~ super intelligence overlords.
Hahaha. Yeah. :(
Who the hell would have guessed that we'd have to deal with not one but two potentially civilization ending threats in our lifetimes? I want off this crazy ride please!
The whole organization structure & how it functions is just not so smart after all. Have management team considered the Lean Methodology with their business objectives?
- The problem that precipitated all this is that they don't have business objectives. They have a "mission." The board of directors of OpenAI aren't beholden to shareholders, and though the staff mocked their statement that allowing the company to be destroyed “would be consistent with the mission” it's actually true.
  
  Appreciate your clerification.

106 comments