Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

There is no hard takeoff

geohot.github.io

Post reply on HN
Print view View original

41–50 of 190 posts

Re: There is no hard takeoff

#41

Post by epiccoleman »

anileated wrote: post #8
ilaksh wrote: post #5

I am all for AI research and integrating more AI use into society. Currently working on tools based on GPT. I think it has incredible potential to help humans. But at the same time, I am sure that AI does not need to have a hard takeoff to be extremely dangerous. It just needs to get a bit smarter and somewhat faster every few months. Within less than a decade we will have systems that output "thoughts" and actions a…

To see how “AI” can be dangerous just look how social media bubble/recommendation algorithms would radicalize people, even with much cruder ML. People tend to miss that it’s not about some model starting to “think” and act all sci-fi evil, just us humans applying powerful tech in irresponsible ways that we either don’t bother to assess due to lack of awareness or assess positively due to a conflict of interest (money…

Right. This has been my take since ChatGPT hit the scene.

I'm not really afraid of Skynet or AM. What worries me is that AI will accelerate the enshittification of everything as it gets baked into everything.

Think of the experience of trying to get any kind of help from a giant company with call centers. You have some legitimate problem or grievance. Any thinking person would agree that a resolution is in order.

But you're not allowed to access a thinking person. What you get instead is some wage slave who has to follow a script. The script doesn't have your situation in it. They don't have some button they can press to solve your problem. There's a gigantic bureaucracy between you and the person who could fix your issue. Eventually you may just give up and mark your issue down as unsolvable.

This is already a realistic situation today, but now imagine that there's a new layer in front of all this where you have to convince an LLM your problem is worth considering. Or imagine the human on the other end has to try to send your request through some AI system.

It's not going to launch all the nukes, or construct nanofactories to make a plague. It's just going to get in your way, serve you garbage, frustrate you, and make the world a slightly worse place.

Maybe in the far future it could be more of a Skynet situation, I dunno. But between then and now there will be plenty of low level annoyance for anyone having to deal with these systems.

I'm not as cynical as this post reads - I think just like the whole internet there are still opportunities for good here, for people's lives to be improved by technology. But our incentives right now sure encourage the worse scenario.

Re: There is no hard takeoff

#42

Post by Hendrikto »

hardware2win wrote: post #38

I will be worried when AI will constantly beat top1 league of legends team in normal environment Chess is way simpler imo.

Funny how the benchmarks for proving ”real intelligence“ change every time there is a breakthrough.

* ”X would demonstrate real intelligence.“

* Some ML model M beats X.

* ”Well, actually, M is not really intelligent. Y would prove real intelligence.“

* repeat

Re: There is no hard takeoff

#43

Post by sidcool »

George Hotz recent talk during Comma Con was pretty dope. He was pointing to a future where Comma would not be a self driving system, but a general purpose robotics one. And his confidence was great too. I know he made a mess out of his twitter stint, but I still feel he's a good engineer/entrepreneur.

Re: There is no hard takeoff

#44

Post by anonymoushn »

neatze wrote: post #6

> Oh wait…every hedge fund bro is already doing this. And most of them aren’t billionaires. The problem is your model needs to include all the computers playing the market, and it also needs to include the other hedge fund bros themselves. This strategy only dominates if you have more compute than the whole market itself, which you don’t. Best part of this article, honestly, I did not expected this, of-course one can…

This is more or less how some multibillion dollar firms work. An issue with the strategy is that of course you cannot just do this, your up-front costs to acquire the data and connectivity are tens of millions of dollars.

Re: There is no hard takeoff

#46

Post by em500 »

mysterydip wrote: post #20

Do we have enough historical weather data to train a decent AI on? Maybe this is already being done and is just a quieter area of research (in the headline sense)? I know for example every time a hurricane comes toward land all the weather places trot out a half dozen different model predictions saying different things. It would be great if there was more confidence in this area for planning.

> Do we have enough historical weather data to train a decent AI on?

I have no particular insights about weather data or models, but I want to note that some prediction problems are just not limited by historical data (as you seem to assume implicitly). Consider the simplest example of a coin toss: the outcome of the next toss in unpredictable no matter how much historical coin toss data you collect and no matter how sophisticated your AI/ML model. (The only thing you can predict is the fraction or amount of heads/tails in a large batch of coin tosses, and you don't need much historical data for that.)

Re: There is no hard takeoff

#47

Post by nwoli »

reedf1 wrote: post #32

He makes a good point. The only real way to "summon the demon" is if AGI is centralized. If each of us has a maximally intelligent AGI in our phone - and it is us as individuals controlling the prompts, then the playing field is levelled. Unstoppable force meets unstoppable force. Gun meets gun. A tale as old as society.

We also already have laws against evil behaviour, no need to introduce AI specific regulation. People can poison the water supply, or make bio weapons already. Yet people don’t because there’s laws and the same will stop people from using AI for evil.

Re: There is no hard takeoff

#48

Post by ilaksh »

moonchrome wrote: post #10
ilaksh wrote: post #5

I am all for AI research and integrating more AI use into society. Currently working on tools based on GPT. I think it has incredible potential to help humans. But at the same time, I am sure that AI does not need to have a hard takeoff to be extremely dangerous. It just needs to get a bit smarter and somewhat faster every few months. Within less than a decade we will have systems that output "thoughts" and actions a…

> It just needs to get a bit smarter and somewhat faster every few months. Within less than a decade we will have systems that output "thoughts" and actions at least dozens of times faster than humans. I don't think this will work because cost of improvement with current method is exponential and we're already at capacity with HW.

It only needs to get somewhat smarter. The hardware, software and AI models will continue to be sped up.

Also there are already many ways that improvements have been made without requiring exponentially more memory.

Re: There is no hard takeoff

#49

Post by mkl »

mysterydip wrote: post #20

Do we have enough historical weather data to train a decent AI on? Maybe this is already being done and is just a quieter area of research (in the headline sense)? I know for example every time a hurricane comes toward land all the weather places trot out a half dozen different model predictions saying different things. It would be great if there was more confidence in this area for planning.

No, not by many orders of magnitude. Even if you had sensors measuring temperature, pressure, humidity, wind velocity, etc. in every cubic metre of atmosphere, weather is chaotic (a tiny variation in starting conditions can lead to completely different outcomes) so you'd still only be able to predict a limited time ahead.

Meteorologists have been working on this problem for a long time, and have thought of and tried every obvious idea (and many more). Weather prediction is literally where modern chaos theory started: https://en.wikipedia.org/wiki/Chaos_theory#History:~:text=Hi...

Re: There is no hard takeoff

#50

Post by threeseed »

scaramanga wrote: post #40

Am I the only one who sees a jumbled stream of barely coherent semi-thoughts and word salad?

The inventor of UML, Grady Booch deals with people like this on Twitter every day.

People seem to think that because ChatGPT appears to be intelligent that somehow we are on the cusp of a new AGI world. And that we need to serious thought leaders like George Hotz and other arrogant tech-bros to rescue us. It's just crypto all over again.

The real thought leaders are going to be those in academia, government etc. Boring policy people who know the subtleties of the real world far better.

Post reply on HN