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

After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

gizmodo.com

Post reply on HN
Print view View original

111–120 of 392 posts

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#111

Post by chpatrick »

crazygringo wrote: post #52

It's so confusing because people keep (intentionally?) conflating two separate ideas of "AI safety". The first is the kind of humdrum ChatGPT safety of, don't swear, don't be sexually explicit, don't provide instructions on how to commit crimes, don't reproduce copyrighted materials, etc. Or preventing self-driving cars from harming pedestrians. This stuff is important but also pretty boring, and by all indications c…

A few years ago we thought we were so far away from AI creating art or having a reasonable conversation.

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#112

Post by rondini »

invig wrote: post #83
broast wrote: post #78

Earlier quoted context omitted.

It's possible we could stumble on AGI without immediately realizing it, and without checks and balances built in, it could already be too late.

The same is true for anything. It's possible we could stumble into a dragon and unless we've built dragon defences it'll be too late.

And if we're hunting for dragons should we wait until we find one and bring it to the castle before we start discussing how to protect people from it?

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#113

Post by bobba27 »

Animats wrote: post #16
temp0826 wrote: post #3

Earlier quoted context omitted.

The only threat is automated blogspam creation. LLMs are not AI (not even close), only marketed as such.

LLMs are generating more and more of journalism. Especially sports and financial coverage, which is taking some numbers and blithering about them.

[flagged]

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#114

Post by Waterluvian »

AI is capital intensive, scaling with user count. That makes any sort of “do the right thing” fundamentally at odds with “make money or at least break even.”

I love the idea but it’s just not going to be a thing. Do not be fooled by what OpenAI says. Pay attention to what they do.

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#115

Post by shrimpx »

From now on we need to actively lash back/keep eyes wide open whenever a SV exec/founder starts spreading their cynical PR about “benefiting humanity”, “connecting humanity”, “opening things up for all of humanity to enjoy”, “making things safe for humanity”, “solving humanity’s greatest threat”, etc.

This was always the case — put a bullshit brand in front of cynical extractive machinery — but in this AI saga it’s reaching mind boggling levels of cynicism, including incorporating patsy nonprofits and holding congressional sessions to peddle sociopathic projective bullshit to our elected lawmakers. (You need to slow down the thing from being made (which is the thing I’m making and working on scaling up as fast as possible lol!))

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#116

Post by foota »

AYBABTME wrote: post #109
crazygringo wrote: post #52

It's so confusing because people keep (intentionally?) conflating two separate ideas of "AI safety". The first is the kind of humdrum ChatGPT safety of, don't swear, don't be sexually explicit, don't provide instructions on how to commit crimes, don't reproduce copyrighted materials, etc. Or preventing self-driving cars from harming pedestrians. This stuff is important but also pretty boring, and by all indications c…

I find it confusing why the first type of AI "safety" is being focused on so much. A reasonable person will know that a tool like AI can spit out none-politically-correct things, and that reasonable person should shrug it off. But we're making a giant deal out of it and dedicating an unreasonable amount of work onto this aspect, at least in my opinion. It's akin to spending 1/4 of a Dewalt drill's development budget…

I think it's a mix of personal opinions, fear of regulation, and fear of public/corporate backlash.

Not to mention, if you're building a chatbot API for someone, no one will use it if customers could accidentally send it into a profanity loop.

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#117

Post by chpatrick »

taf2 wrote: post #38
golol wrote: post #15

Look, there is ASI/AGI safety, and "please don't swear and scam and misinform, and be unbiased" safety. ASI/AGI safety is only a concern right now if you believe in foom , i.e. that an AGI would rapidly self improve to an ASI and extinction level threat. I think it is absurd to believe that because AI right now is compute limited and any first AGI will only be able to run on large data centers. The latter kind of saf…

I just view the current gpt llm as a better search engine. It’s not self learning. It’s stateless when queried - so really the concern is we can create content on demand and create responses that intuitively make the llm very useful tool. Until it’s self learning I’m only concerned with how humans would use it just like any other tool in the hands of humans it can be both super useful and super destructive.

If they figure out how to give an LLM infinite/really long context length, it will basically become self-learning.

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#118

Post by Apocryphon »

hackerlight wrote: post #74
nradov wrote: post #39

Earlier quoted context omitted.

Help us out here. Let's say that known bad actor Vladimir Putin gets an AGI. How does that allow him to take over the world despite many other countries having nuclear weapons? Take us through it step by step, and no hand waving please.

In the short term, doing what he's been doing but 10x more effectively. Waging an information war on Western democracies via social media that causes them to collapse due to infighting and division. Also steering internal attitudes of the voter base to a desirable direction.

I was actually going to suggest that. If we are to draw on what LLM can currently do, we know that it can imitate human writing pretty well and pontificate at length at a variety of subjects in different formats.

So if you wanted a tool to generate a lot of blogspam, clickbait, wiki hoaxes, and fake news, an LLM could probably expedite the work the infamous Internet Research Agency is purported to do. Of course, there’s still time remaining for the kinks to get worked out and it fully sounds convincing and like the natural language from a native speaker, etc.

And I don’t think existing LLMs are able to know where to deploy these propaganda attacks yet. That really requires human social analysis. So people are going to have to be involved in every step of the loop.

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#119

Post by spunker540 »

DalasNoin wrote: post #75
needadvicebadly wrote: post #66

Earlier quoted context omitted.

I don't mean to be facetious in any way, but how does AGI/ASI protect itself from humans simply pulling the plug?

Seems like it was pretty hard to pull the plug on sam? It will prevent you from pulling the plug by being smarter than you, spreading to more systems, making you dependent on it, persuasion,...

Neat idea, so it will conform like most humans in society and try to create more value than it destroys to avoid persecution and jail time? Sounds like our shared societal values will keep it in check then the same way they kept the openai board in check.
Post reply on HN