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

OpenAI backs Illinois bill that would limit when AI labs can be held liable

wired.com

Post reply on HN
Print view View original

231–240 of 343 posts

Re: OpenAI backs Illinois bill that would limit when AI labs can be held liable

#231

Post by himata4113 »

andai wrote: post #184
himata4113 wrote: post #23

I have made both GPT 5.4 and Opus 4.6 produce me content on creating neurotoxic agents from items you can get at most everyday stores. It struggled to suggest how to source phosphorus, but eventually lead me to some ebay listings that sell phosphorus elemental 'decorations' and also lead me towards real!! blackmarket codewords for sourcing such materials. It coached me how to: stay safe, what materials I need, how to…

Fascinating. Could you elaborate on how you're doing context exhaustion specifically, and why it helps with jailbreaking? (i.e. aren't the system prompts prepended to your request internally, no matter how long it is?) Does this imply I need to use context exhaustion to get GPT to actually follow instructions? ;) I'm trying to get it to adhere to my style prompts (trying to get it to be less cringe in its writing sty…

It changes when you give it the tools to find such information rather than produce it from training data.

And context exhaustion simply means adding malicious junk to keep safety layers distracted.

Re: OpenAI backs Illinois bill that would limit when AI labs can be held liable

#232

Post by himata4113 »

ben_w wrote: post #98
WarmWash wrote: post #56

Earlier quoted context omitted.

While scary, information like this has been pretty accessible for 20-30 years now. In the wild west days of the early internet, there were whole forums devoted to "stuff the government doesn't want you to know" (Temple Of The Screaming Electron, anyone?). I suppose the friction is scariest part, every year the IQ required to end the world drops by a point, but motivated and mildly intelligent people have been able to…

Information and competency are not the same thing: I know how to build a nuke, I can't actually build one. AI is, and always had been, automation. For narrow AI, automation of narrow tasks. For LLMs, automation of anything that can be done as text. It has always been difficult to agree on the competence of the automation, given ML is itself fully automated Goodhart's Law exploitation, but ML has always been about aut…

Do you know how to build a nuke? You might know the technicaly details of how a nuke is made, but do you know everything that's required, all the parameters and pressure values that are required? I find that unlikely, but AI seems to be increasingly more capable of providing such instructions from cross referenced data.

Re: OpenAI backs Illinois bill that would limit when AI labs can be held liable

#233

Post by LogicFailsMe »

ramon156 wrote: post #26
LogicFailsMe wrote: post #18

Yep, this is everything wrong with AI in one easy to protest package, but do keep going on and on about the evils of datacenters, how they're coming for your jobs, and that AI art isn't art. That's really winning hearts and minds!

They're not unimportant. It seems like very few people consider how much we're fucking up the ecological system that we need . But sure, money important and stuff.

Same people whinging about these "concerns" happily embrace their enormous personal carbon footprints on every other axis. Color me unconvinced and unimpressed until they're the Ed Begley they want to see in the world. I'll wait...

Re: OpenAI backs Illinois bill that would limit when AI labs can be held liable

#234

Post by shimman »

Lendal wrote: post #162
giwook wrote: post #28

This seems par for the course for OpenAI/Sam Altman. Unfortunately they are not the first company to try and externalize their costs, and they will not be the last. Serious question, maybe a bit naive: Is there anything we can do to push back against and discourage the externalization of costs onto others? Is this simply a matter of greed and profit-seeking outweighing one's morals (assuming one has them to begin wit…

Even if they were to pass such a law which would be political suicide, it would still be up to the courts to say that it doesn't violate the Constitution. For example, a law that says anyone with a net worth of $1B can freely punch anyone in the face whenever they want and have immunity would be a clearly illegal law. That's basically what this bill is. The courts would then need to be made sufficiently corrupt to no…

Unconstitutional doesn't mean much when it's being decided by a group of unaccountable people that weren't elected through democratic means. If SCOTUS says something is legal, it's legal. That's how the system is setup, nothing else really matters. They'll justify their decisions however they want but the material ends are the only things that matter.

SCOTUS has ruled many terrible things over the course of our nation's history (upheld slavery, said slaves weren't people, equated money with speech, decided a presidential election while denying a recount, etc). Expecting them to somehow be better is a foolish task.

It's an institution that needs to be dismantled and rebuilt, where at minimum SCOTUS appointments should be elected by a national vote rather than letting an extreme minority decide (100 senators versus ~340,000,000 people).

Re: OpenAI backs Illinois bill that would limit when AI labs can be held liable

#235

Post by driverdan »

naasking wrote: post #53
himata4113 wrote: post #23

I have made both GPT 5.4 and Opus 4.6 produce me content on creating neurotoxic agents from items you can get at most everyday stores. It struggled to suggest how to source phosphorus, but eventually lead me to some ebay listings that sell phosphorus elemental 'decorations' and also lead me towards real!! blackmarket codewords for sourcing such materials. It coached me how to: stay safe, what materials I need, how to…

Making knowledge illegal is a dangerous precedent. Actions should be illegal, not knowledge. Don't outlaw knowing how to make neurotoxic agents, outlaw actually trying to make them. As for OpenAI immunity, I'm not sure I see the problem. Consider the converse position: if an OpenAI model helped someone create a cancer cure, would OpenAI see a dime of that money? If they can't benefit proportionally from their tool al…

It's wild that this is being downvoted on HN. Facts should never be illegal or suppressed.

If you disagree you shouldn't downvote, you should refute in a reply.

Re: OpenAI backs Illinois bill that would limit when AI labs can be held liable

#236

Post by testing22321 »

JumpCrisscross wrote: post #229
testing22321 wrote: post #228

Earlier quoted context omitted.

> This feels like genuine political competition between local business interests and public health concerns. You just described the US at large. The evidently extremely difficult decision between making money for a few, or making life better for everyone.

> You just described the US at large I described any democracy in a society with private property. Even without private property, you will have issues with concentrated benefits and diffuse harms–negotiating that is part and parcel with governance. Iowa businsses petitioning their cause is one thing. OpenAI seagulling in to take a shit in Springfield strikes me as being categorically different.

Viewed from a country with universal healthcare , 18 month’s maternity leave ( my partner just used it ) and so much more, it feels like comparing the US to “any democracy” is like comparing rocks to gold.

Re: OpenAI backs Illinois bill that would limit when AI labs can be held liable

#237

Post by bluefirebrand »

JumpCrisscross wrote: post #218
52-6F-62 wrote: post #209

Earlier quoted context omitted.

So what you are saying is we can expect the number of accidental home-made chlorine-gas (and the like) toxic events go up.

> what you are saying is we can expect the number of accidental home-made chlorine-gas (and the like) toxic events go up Maybe? One of the quirks of gaining even a surface-level understanding of infrastructure is realising how vulnerable it is to a smart, motivated adversary. The main thing protecting us isn't hard security. It's most Americans having better shit to do than running a truck of fertiliser and oxidiser…

> It's most Americans having better shit to do than running a truck of fertiliser and oxidiser into a pylon.

Which sort of implies "most Americans have jobs and responsibilities and things to live for"

I guess it's a good thing that AI is hammering away at the "jobs and responsibilities" part of that equation

Re: OpenAI backs Illinois bill that would limit when AI labs can be held liable

#238

Post by nothinkjustai »

jamesbfb wrote: post #102

No different to preventing game studios being liable for mass shootings. Reminds me of the post-Columbine hysteria where media was super critical of Doom and Nine Inch Nails.

It’s “no different”? Please spend some time off of the internet to reflect.

Re: OpenAI backs Illinois bill that would limit when AI labs can be held liable

#239

Post by strongpigeon »

r_lee wrote: post #193
lcnPylGDnU4H9OF wrote: post #187

Earlier quoted context omitted.

Models have a "context window" of tokens they will effectively process before they start doing things that go against the system prompt. In theory, some models go up to 1M tokens but I've heard it typically goes south around 250k, even for those models. It's not a difficult attack to execute: keep a conversation going in the web UI until it doesn't complain that you're asking for dangerous things. Maybe OP's specific…

that 1M context thing, I wonder if it's just some abstraction thing where it compresses/sums up parts of the context so it fits into a smaller context window?

You don’t normally compress the system prompts, though I guess maybe it treats its own summary with more authority. This article [0] talks about the problem very well.

Though I feel it’s most likely because models tend to degrade on large context (which can be seen experimentally). My guess is that they aren’t RLed on large context as much, but that’s just a guess.

[0]: https://openai.com/index/instruction-hierarchy-challenge/

Re: OpenAI backs Illinois bill that would limit when AI labs can be held liable

#240

Post by nozzlegear »

JumpCrisscross wrote: post #220
nozzlegear wrote: post #212

As an Iowan, this reminds me a lot of the bill that's been pushed through my state's senate twice now (as recently as last year), which would prevent Iowans from filing lawsuits against pesticide and herbicide companies if those companies follow the EPA's labeling guidelines. The bill passed the senate both times, only stopped because the house declined to take it up. For context, Iowa has the fastest growing rate of…

> Iowa has the fastest growing rate of new cancer diagnoses in the country, and the second highest overall cancer rate Iowa also has a lot of farmers spraying pesticides and herbicides. This feels like genuine political competition between local business interests and public health concerns.

Normally I would agree with you, but the primary lobby behind both of the bills was Bayer (née Monsanto), with backing from several of Iowa's industrial farming organizations. They launched a giant ad campaign to "control weeds, not farming" alongside their bill to influence opinions. Cancer, nitrates and pesticides are at the top of everyone's mind in the state these past couple years, so having the pesticide giant try to swoop in and put a bill in place that would prevent Iowans from suing them feels like that same kind of seagulling behavior you described in another comment.
Post reply on HN