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

Pause Giant AI Experiments: An Open Letter

futureoflife.org

Post reply on HN
Print view View original

341–350 of 1001 posts

Re: Pause Giant AI Experiments: An Open Letter

#341

Post by echelon »

comment_ran wrote: post #283

"The Dark Forest" is a science fiction novel by Liu Cixin, which introduces a concept called "suspicion."(“猜疑链”) In this concept, the interaction between civilizations is viewed as an ongoing game of suspicion, where each civilization is worried that the other civilization may pose a threat to it. This kind of suspicion makes it difficult for civilizations to establish trust, thereby making the universe a dark forest…

Except the concern isn't genuine here. Some of the signatories have AI companies and interests. They want to slow the leaders down so that they can catch up. It's disingenuous.

Re: Pause Giant AI Experiments: An Open Letter

#342

Post by Quarrelsome »

c7DJTLrn wrote: post #278
Quarrel wrote: post #257

Earlier quoted context omitted.

Forget countries? How will you stop me? I have models on my local computer. I can and will use them. They will get better. You don't put tech genie's back in their bottles, that's not how it works.

>I have models on my local computer. I can and will use them. They will get better. Well no, not really. You're just consuming the result of millions of dollars of investment. You're not going to be able to improve it in any significant way with limited resources and knowledge.

you can deploy it in novel situations and release it unethically. Are we not already at the point where we can effectively simulate social media chatter and thus people or social pressure with publicly available tools?

If the API gate is broken then the cat is out of the bag.

Re: Pause Giant AI Experiments: An Open Letter

#343

Post by dmarchand90 »

zug_zug wrote: post #69

I understand that some people find this a hopeless gesture. But I think anybody who thinks "losing jobs" or "China" is a bad outcome from AI is hopelessly lost. The upside of AI if done right - nobody ever having to work again The downside of AI if done wrong - end of humanity It doesn't take a rocket scientist to determine we've made AIs go from thinking like an ant to thinking like a 6-year-old in a few years. Anyb…

Maybe not as fast as you'd think. Imagine if you dropped a human off in a time machine with a bunch of chimps...

Ultra high iq people do better on average, but correlation with success drops down after 120 or so

I'm also very skeptical that there will be an exponential takeoff as the energy costs are immense and increases proportionately with intelligence. So until it solves fusion power for us I'm not super worried

Re: Pause Giant AI Experiments: An Open Letter

#344

Post by mclightning »

kleer001 wrote: post #339
comment_ran wrote: post #283

"The Dark Forest" is a science fiction novel by Liu Cixin, which introduces a concept called "suspicion."(“猜疑链”) In this concept, the interaction between civilizations is viewed as an ongoing game of suspicion, where each civilization is worried that the other civilization may pose a threat to it. This kind of suspicion makes it difficult for civilizations to establish trust, thereby making the universe a dark forest…

Hmm, sounds like parties not fully grokking the optimal (not thinly logical) solution to the iterated prisoners' dilemma. I saw the same thought process is the otherwise very nice novel "Time's Children". Good for drama, but not thought through.

I have so many questions >> what does it mean to fully grok the optimal? >> what is the iterated prisoners dilemma?

Re: Pause Giant AI Experiments: An Open Letter

#346

Post by danaris »

labrador wrote: post #301
tgv wrote: post #281

Earlier quoted context omitted.

> Elon thought AI would enslave us, but turns out it empowers us. I'm not sure how you reached that conclusion. GPT3 is 2 years old. It has ample time to enslave us. I'm also still waiting for the "empowering" bit. Haven't really noticed it, I must say. Or are you referring to a few code completion tricks?

If you don't feel empowered by AI, imho you're doing it wrong and will probably be one of those left behind to be enslaved. I'm confident in my ability to stay on top, with a help from my own AI.

Or you're going to find that suddenly the centralized, proprietary AI from a for-profit company that you rely on...

a) Puts up a paywall requiring an active subscription to continue use

b) Shuts down because not enough people subscribed

c) Shuts down the services you use because they want to move on to some new shiny

d) Takes all the data you feed it and creates a high-profile, well-funded competitor to whatever you're working on

e) Starts poisoning your output once it's clear that you're competing with them

f) Starts poisoning your output to create subtle propaganda for their own products or political beliefs

g) Starts poisoning your output to create subtle propaganda for anyone who will pay them

...And those are just the failure modes I can think of in a couple of minutes off the top of my head.

Re: Pause Giant AI Experiments: An Open Letter

#348

Post by dmarchand90 »

adamsmith143 wrote: post #316
thomastjeffery wrote: post #302

The dumb criticize the blind. What an absurd situation! How did we get here? Here are the steps: 1. Large Language Models have been presented as "AI", which personifies them instead of describing how they work. 2. Goals for LLM development were set for the personified attributes, and not the actual functionally of the real thing. OpenAI brags about how GPT4 scores at human tests: as if that has any bearing on the mod…

At no point in your rambling, incoherent comment were you even close to anything that could be considered a rational thought. Everyone on this site is now dumber for having read it. May God have mercy on your soul.

https://news.ycombinator.com/newsguidelines.html

Re: Pause Giant AI Experiments: An Open Letter

#350

Post by braingenious »

>AI research and development should be refocused on making today's powerful, state-of-the-art systems more accurate, safe, interpretable, transparent, robust, aligned, trustworthy, and loyal.

LMAO.

“AI should have… waves at a laundry list of nice sounding adjectives that mean different things to every single signatory in order to accomplish… Humanity [can enjoy] a flourishing future with AI. This will take… six months.”

Six months doesn’t sound long enough to guarantee AI safety, but it sounds about long enough for Elmo to train up and release his anti-wokeness LLM (whatever that means), which he has been pretty vocal about being important.

Post reply on HN