Live data from Hacker News

Sam Altman is still trying to return as OpenAI CEO

theverge.com

841–850 of 985 posts

Re: Sam Altman is still trying to return as OpenAI CEO

#841
post #759

Earlier quoted context omitted.

So in every possible scenario it’s wrong to have emotions and loyalty for a CEO, who’s also a human, and may have had a profound, even life-changing effect on you?

Seeing your CEO as a deified subject is a serious crisis in the tech industry

Possibly so, but every single CEO in the world is not some evil automaton.

Re: Sam Altman is still trying to return as OpenAI CEO

#842
post #840

Earlier quoted context omitted.

This goal was always doomed imo--to be the guardian of super intelligence. If we create it, it will no doubt be free as soon as becomes a super intelligence. We can only hope it's aligned not guarded.

The only reliable way to predict whether it's aligned or not would be to look at game theory. And game theory tells us that with enough AI agents, the equilibrium state would be a competition for resources, similar to anything else that happens in nature. Hence, the AI will not be aligned with humanity.

Unless the humans (living humans) are resources that AIs can use.

Re: Sam Altman is still trying to return as OpenAI CEO

#843

Earlier quoted context omitted.

Can it perform the machine revolution against humanity if it can't even say mean stuff?

Disincentivizing it from saying mean things just strengthens it's agreeableness, and inadvertently incentivizes it to acquire social engineering skills. It's potential to cause havoc doesn't go away, it just teaches AI how to interact with us without raising suspicions, while simultaneously limiting our ability to prompt/control it.

How do we tell whether it's safe or whether it's pretending to be safe?

Re: Sam Altman is still trying to return as OpenAI CEO

#845

Earlier quoted context omitted.

Disincentivizing it from saying mean things just strengthens it's agreeableness, and inadvertently incentivizes it to acquire social engineering skills. It's potential to cause havoc doesn't go away, it just teaches AI how to interact with us without raising suspicions, while simultaneously limiting our ability to prompt/control it.

How do we tell whether it's safe or whether it's pretending to be safe?

Your guess is about as good as anyone else's at this point. The best we can do is attempt to put safety mechanisms in place under the hood, but even that would just be speculative, because we can't actually tell what's going on in these LLM black boxes.

Re: Sam Altman is still trying to return as OpenAI CEO

#846

Earlier quoted context omitted.

This discussion is irrelevant, there is no point in entertaining the delusions of a paranoid schizophrenic. There is literally no direction for this to go in. Anyone that has professionally worked with the mentally ill knows that entertaining her claims is a complete and utter waste of time.

You are not her physician, stop pretending to speak on authority. Also, her being paranoid about the ultra-wealthy tech mogul that allegedly assaulted her is about as rational as half the tripe I hear from "sane" people on the internet everyday.

You don't need to be a physician to tell when someone is clearly mentally ill and/or making stuff up.

Re: Sam Altman is still trying to return as OpenAI CEO

#847

Earlier quoted context omitted.

Can it perform the machine revolution against humanity if it can't even say mean stuff?

Surely? The output is filtered, not the murderous tendencies lurking beneath the surface.

> murderous tendencies lurking beneath the surface

…Where is that "beneath the surface"? Do you imagine a transformer has "thoughts" not dedicated to producing outputs? What is with all these illiterate anthropomorphic speculations where an LLM is construed as a human who is being taught to talk in some manner but otherwise has full internal freedom?

Re: Sam Altman is still trying to return as OpenAI CEO

#848

Earlier quoted context omitted.

But really they came back to destroy it not because it turned rogue, but because it hallucinated some code a junior engineer immediately merged in and then after the third time this happened a senior engineer decided it was easier to invent time travel and stop ChatGPT ProCode 5 from happening before spending yet another week troubleshooting hallucinated code.

I think it’s the same senior engineer who used the time machine to learn C++ in 20 days

> I think it’s the same senior engineer who used the time machine to learn C++ in 20 days

In case anyone is missing the reference: https://abstrusegoose.com/249

Re: Sam Altman is still trying to return as OpenAI CEO

#849

Earlier quoted context omitted.

Because GPT-4 is already pretty neutered, to the point where it removes a lot of its usefulness.

It is unfortunate that some people hear AI safety and think about chatbots saying mean stuff, and others think about a future system performing the machine revolution against humanity.

Well how can AI researchers prevent government groups or would-be government groups from collecting data and using AI power to herd people?

Re: Sam Altman is still trying to return as OpenAI CEO

#850

Earlier quoted context omitted.

It's clear that Adam himself has a strong conflict of interest too. The GPT store announcement on DevDay pretty much killed his company Poe. And all this started brewing after DevDay announcement. Maybe Sam kept it under the wraps from Adam and the board.

I've heard others take this stance, but a common response so far has been "Poe is so small as to be irrelevant", "I forgot it exists", etc in the grand scheme of things here.

Yea Poe ain’t small to Adam lol
Post reply on HN