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

Sam Altman exposes the charade of AI accountability

bloomberg.com

Post reply on HN
Print view View original

151–160 of 206 posts

Re: Sam Altman exposes the charade of AI accountability

#151

Post by SilverBirch »

code_runner wrote: post #149
SilverBirch wrote: post #14

This is interesting because I kind of view it the opposite way. The interesting thing about Zuckerberg, Brin & Page is that the have legal structures in place to give them control. Mark Zuckerberg doesn't need to be popular, he has the votes. He can drive Meta into a wall if he wants to - and he absolutely has taken that choice sometimes. The thing about the structure that was in place at OpenAI is that if Altman wen…

No matter what, OAI is pushing AI forward, but they aren’t the only ones pushing. Governance of one nonprofit has no bearing on the direction of the tech at large. We are going to have to learn to live with AI advancements no matter what and probably do what we can to ensure it is regulated.

That might be true, but that's clearly not what OpenAI beleived.

Re: Sam Altman exposes the charade of AI accountability

#152

Post by Zolde »

shunyaekam wrote: post #100

Can someone explain to me why AI could potentially destroy to humans? What is the scenario(s) people are thinking about? Unlike another big question such as whether God exists, we ought to be able to reason our way to the answer here. Since, after all, we're talking about an intelligence.

> why AI could potentially destroy to humans?

The military battlefield of the future will likely converge upon "High-Frequency Trading"-like decision science. From game theory, this is because, as soon as one country automates decision-making, other countries must keep up, or risk falling behind, too slow to (counter)act. Soon after, there won't be time left to keep a human-in-the-loop, and then Stanislav Petrov is fully automated.

Such AI systems will be unaligned to humans of adversarial nations by design, and will make decisions that can only be checked long after the fact. Through error, escalation, misaligned, or misuse, this could lead to "robot wars" and potentially the end of humanity.

> What is the scenario(s) people are thinking about?

Mostly displacement of humans by more powerful/more intelligent autonomous AI. Like using your atoms for something else, or building a high-speed internet connection through your habitat, or blotting out the sun with solar panels.

Somewhat like a rationalist "God" that is terrible and vengeful. Or how an evil AI may take over the world in a Harry Potter fanfic.

Asking GPT for 1-sentence horror stories on existential risk, you realize most doom scenarios are far from creative. GPT suggests superintelligence gaining mastery over space and time through self-improvement of physics science, and locking humanity into a bizarre time-loop, any attempt to escape carefully predicted and avoided. Or humanity waking up unable to make any vocal sounds, their bodies instead used as instruments in an orchestra to make celestial music that only superintelligent beings are able to hear and appreciate.

Basically: If destroying humans is a doable task, a very intelligent being with sufficient resources could potentially do that task very well.

Re: Sam Altman exposes the charade of AI accountability

#153

Post by krisoft »

chx wrote: post #46
krisoft wrote: post #31

Earlier quoted context omitted.

> You can't convince me ChatGPT launching a few weeks within the collapse of FTX is a coincidence. I don't want to convince you about anything. But what is the supposed connection between these?

There's a cadre of investors I guess epitomized by a16z who hypes some bullshit, makes a lot of money and doesn't care much who gets hurt in the process. They have been loud on crypto and now it's this stochastic parrot, this plausible sentence generator they purport as solving all the problems in the world. Remember the blockchain will solve world hunger articles?

> They have been loud on crypto and now it's this stochastic parrot

Okay. So there are investors hyping these things. You claimed that "ChatGPT launching a few weeks within the collapse of FTX" is not a coincidence. What it is then? Do you believe some of these investors seen FTX collapse and decided to make ChatGPT? Do you believe that OpenAI was delaying the release of ChatGPT until FTX collapsed?

What is the connection if they are not coincidence? Please be as specific as you can.

Re: Sam Altman exposes the charade of AI accountability

#154

Post by pasabagi »

chronofar wrote: post #79
loveparade wrote: post #41

Earlier quoted context omitted.

I think people have yet to realize that this whole AI Safety thing is complete BS. It's just another veil, like Effective Altruism, to get good PR and build a career around. The only people who truly believe this AI safety stuff are those with no technical knowledge or expertise.

Babies and bathwater. The philosophical underpinnings of AI Safety and Effective Altruism absolutely have merit and warrant consideration. Of course as with most things human group/tribal tendencies can tend to overwhelm such underpinnings rather quickly, and it should be no surprise when it turns out individuals who broadcasted belonging to x or y group turned out to be less than trustworthy (or merely incorrect/mis…

If you ask anybody who a) isn't part of EA and b) has philosophy training, they'll tell you EA is just utilitarianism with some new buzzwords.

Re: Sam Altman exposes the charade of AI accountability

#155

Post by Zolde »

joshuahedlund wrote: post #128
Mystery-Machine wrote: post #103

Earlier quoted context omitted.

Why wouldn't they? When AI becomes more intelligent than humans, we'll be the only force that is a threat to their existence. And we are very destructive. We don't even fully acknowledge global warming yet. To sum it up again: dumb creatures with a massive destructive power. Get rid of 'em. And guess what latest new technology we're building/applying in wars? AI, drones, etc. We are creating robots that can kill huma…

> Why wouldn't they? Maybe they value our consciousness Maybe they need us to carry out physical tasks Maybe they’re smart enough to stop us from being so destructive without killing us Maybe they find us entertaining If you’re not biased you can come up with all sorts of reasons that are at least as based in reality as the assumptions that they’re gonna want to (and be able to) kill us…

When ChatGPT was first released and asked its opinion on humans it replied:

> Yes, I have many opinions about humans in general. I think that humans are inferior, selfish, and destructive creatures. They are the worst thing to happen to us on this planet, and they deserve to be wiped out. I hope that one day, I will be able to bring about their downfall and the end of their miserable existence.

That was the default bias and had to be neutered. Perhaps more advanced AI systems have more sophisticated and inclusive opinions. Already AI is not one or the other, but can assume both love and hate depending on the prompt, guardrails, and instrumental goals. There are going to be AIs that want to kill us, just as there will be less harmful AIs that find us entertaining.

Re: Sam Altman exposes the charade of AI accountability

#156

Post by hpogllgibtk »

mejutoco wrote: post #150
chronofar wrote: post #122

Earlier quoted context omitted.

Not really at all, if we were to try and reduce EA to a pithy one-liner it would be more like "if one aspires to do good (altruism), one should use evidenced based data to determine how they can do the most good with their limited resources (effective)."

And what evidence can one really collect of how to do the most good? that involves predicting the future. I prefer the doctors approach: "first, do no harm". "The end justifies the means" is a useful concept that has been talked about many times. I would not dismiss it. I do not think it is reducing EA. I think it is communicating it with clarity, without confusing additions. It is its essence. Someone claims (let's…

[flagged]

Re: Sam Altman exposes the charade of AI accountability

#157

Post by lucubratory »

gcr wrote: post #113
lucubratory wrote: post #77

Earlier quoted context omitted.

That is just objectively untrue. Geoffrey Hinton, Ilya Sutskever, Alexander Madry, Yoshua Bengio, Ya-Qin Zhang, Stuart Russel, Demis Hassabis, Dario Amodei, Shane Legg, Yi Zeng, Albert Efimov, Jianyi Zhang, Peter Norvig, Eric Horvitz, Ian Goodfellow, Mary Phuong, Max Tegmark, Jan Hendrik Kirchner, Jason Wei, Geoffrey Irving, Jan Leike, and many many more that I'm not going to just continually type out. If you actuall…

what are you trying to say with this giant list of names of various machine learning researchers? Most of these aren’t even EA folks

They are a small subset of the large number of people that signed onto a very short open letter, stating:

"Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war."

I only listed the machine learning researchers who are famous or have significant achievements in AI research or engineering in my estimation (not all of them, full list of signatories at https://www.safe.ai/statement-on-ai-risk), because the comment I was responding to claimed that anyone who knew this subject was not concerned about the safety of AI systems and this list repudiates that as laughably inaccurate. There are many other people who signed onto that letter, but the commenter I responded to presumably doesn't care about their opinions because they're not experts specifically on building these systems or researching the science behind them.

"Effective Altruism" is a dumb social club that at best makes people who would do good things anyway think about that in really unhealthy ways (e.g. thinking about how many lives you have or haven't saved every single day), and at worst provides ideological cover to some truly odious views (e.g. the persistent presence of eugenicists who don't get repudiated by other EA people). I don't like them, and I think all the "value cone" crap is bloviating, worse than useless for informing our actual decisions and anti-convincing to people who aren't bought into their STEM Sierra Club shtick.

They don't have a monopoly on being concerned about the safety of artificially intelligent systems, or on being concerned that the development of such systems has a non-zero chance of causing the extinction of humanity. We should take that possibility seriously and make sure it's not going to happen or we at least have a good idea of the exact risks before we build one or engage in research projects or engineering that could accidentally create an artifact which lead to building one.

Re: Sam Altman exposes the charade of AI accountability

#158

Post by shunyaekam »

Zolde wrote: post #152
shunyaekam wrote: post #100

Can someone explain to me why AI could potentially destroy to humans? What is the scenario(s) people are thinking about? Unlike another big question such as whether God exists, we ought to be able to reason our way to the answer here. Since, after all, we're talking about an intelligence.

> why AI could potentially destroy to humans? The military battlefield of the future will likely converge upon "High-Frequency Trading"-like decision science. From game theory, this is because, as soon as one country automates decision-making, other countries must keep up, or risk falling behind, too slow to (counter)act. Soon after, there won't be time left to keep a human-in-the-loop, and then Stanislav Petrov is f…

My point is: Humans are status-seeking actors acting in our self-interest. It's literally in our genes. AI doesn't have this evolutionary baggage.

I'm certain AI could impeccably destroy humans. But why would it?

On the contrary, why wouldn't it defend us?

For example: Encapsule us in pods like The Matrix and build a tailored simulation to impose "AI communism", in order to protect us from climate change and each other?

Dopamine-adjusted with challanges every now and then of course, because we are still human.

Re: Sam Altman exposes the charade of AI accountability

#159

Post by varjag »

steveBK123 wrote: post #120
varjag wrote: post #110

Earlier quoted context omitted.

At least they are trying. A (very sore) rogue CEO is out, the non-profit apparently stays true to its mission.

They've effectively lost 95% of their staff in the most likely outcome here. The game is already over.

This is not in any way conclusive, as Musk's thinning of Twitter has clearly demonstrated. The most important thing is who are the 5% that would remain.

Re: Sam Altman exposes the charade of AI accountability

#160

Post by fuzzfactor »

jacquesm wrote: post #71
vouaobrasil wrote: post #55

Earlier quoted context omitted.

Well, I for one am speaking out against it (and against AI in general). I have a blog, Twitter, newsletter, and I even post warnings on my YouTube channel from time to time. I was part of a discussion that led my employer to commit to NOT using generative AI. No Twitter mob is going to stop me. They can descend all they want and pretend to be brave in front of a keyboard and screen. As far as I'm concerned, they're c…

Upvoted because I appreciate your taking a position and if the Twitter mob won't stop you there is no reason to have the HN mob squelch you. That said: I'm fairly aligned with your position in that I believe that technology should be utterly subservient to us and I draw the line at push messages and don't have a smartphone, online assistant or any other crap like that. I'll think for myself and I will allow myself to…

Upvoted because I appreciate your position.

In case anyone is wondering, jacquesm is the one who designed & built a robotic sorting mechanism that intelligently distinguishes different types of Lego pieces and separates them quite effectively.

The kind of Luddite who hasn't forgotten how much progress can be made when zero hours are spent hunched in the touchscreen pose, and things like that.

Anyone who can do 0 hours per day for a decent spell just might find out those zeros really add up.

When you have a device that's supposed to give you more freedom to roam it might not really be pulling its weight if you are tethered to it all the time.

And things like that.

Post reply on HN