Hacker Newsnew | past | comments | ask | show | jobs | submit | combobyte's commentslogin

It feels like there's should be some sort of "demonstrable intent to do harm" statute on the books for situations like this.

If you have a genuinely-held belief that AI systems will end life on earth and you willfully continuing working on them anyway, then, like... you probably shouldn't be allowed to walk free in civil society anymore, right?


This is basically already the case in the flourishing indie space, but for different reasons.

The only developers chasing cutting-edge hardware features are the AAA studios and I honestly think a lot of them would just die before successfully relearning how to make games that play well instead of just look pretty.


I'd argue we've been in an "everything-has-an-API" world for a long time now — it's just that discoverability of said APIs is still crap.

And since LLMs are apparently good at circumventing the absence of an API, there's not much incentive to add them now. APIs are for humans. LLMs just break through all the captchas and anti-bot measures.

Humans do this too.

Yes, but for humans it creates friction. Seeing a captcha makes me think twice and thrice if I really want to visit that site so badly that I'll endure the suckage. LLMs don't care, for them the friction doesn't exist.

It must be tuned for rage-based engagement because Claude only ever uses the most obnoxious Claudisms on me despite constant reminders to knock it off.

When was the last time you saw an ad for cigarettes?

Those second-order effects are easy to mitigate with sensible public policy. And it's not like we have to invent the solution from scratch — we've done it before, very effectively.


This is the correct reply to the constant "but whatabout" replies, that inevitably come up whenever you suggest regulation. "You've regulated bad thing X, but then they'll do Y!" So regulate Y! It's not like legislators only get one crack at a law. We should elect a government willing to move fast and iterate when loopholes are uncovered.


I mean if you're arguing an edge case you've already accepted the central point, so it's not like a response needs to be made. We handle edge cases.


The thing is that US Congress can't make laws. It just can't. It's nonfunctional. When it does that's the rare exception, and it better be perfect the first time because you won't get so lucky twice.


https://legiscan.com/US/legislation?status=passed

Seems like they can, or you mean they can't 'make laws' that you personally care about?


IMO this link proves the poster’s point. We are passing 1-2 laws a week. So far the 119th congress has passed 104 laws

100 years ago, the 69th congress passed over 1500 laws, plus hundreds of continuing resolutions, treaties, etc https://www.loc.gov/collections/united-states-statutes-at-la...


Cloudflare block page - I'm a bot. What does it say?


> Those second-order effects are easy to mitigate with sensible public policy.

Regrettably, sensible public policy seems to be in remarkably short supply these days.


> When was the last time you saw an ad for cigarettes?

Today at the gas station on the outside of the pump and the store door.


I was at a gas station today that sells tobacco. No ads. Its because even displaying tobacco products, let alone advertising them, is not allowed in that jurisdiction.

I'll bet you didn't see any cartoon camels or doctors recomending a brand though, because even where tobacco ads are legal, many tobacco ads are no longer legal.

Whether you see ads, and what kind of ads you see is highly regulated. That's the whole point,


> I was at a gas station today that sells tobacco. No ads. Its because even displaying tobacco products, let alone advertising them, is not allowed.

We clearly don't live in the same states.


Sorry for the ninja edit about jurisdictions.

I live in a different country since we don't have states here.


after the first fifty years? :)

biological organisms develop addiction much faster than political processes


Do you watch movies?


Being embodied is not the important barrier to running experiments. It's having access to a body of resources (i.e. funding and infrastructure).


For some research funding is mostly the salaries of the people doing the research.


What research works that way?


large parts of Computer science, mathematics, theoretical physics, economics, some humanities research and I'm probably missing a ton.


So applied math, math, applied math, applied math, and maybe some others.

That makes sense, although I'd argue that at least in the realm of HEP theoretical physics has extraordinarily expensive kit compared to what scientists make. See: The LHC.


Reducing all these fields to mathematics is incredibly reductive. I don't think it's in good faith.


I think it's telling that "Researchers cost more than any other aspect of research" seems to involve primarily math as inputs and outputs. Like I said the obvious outlier is theoretical physics because that tends to involve significant resources both to explore (supercomputers) and verify (colliders, telescopes, interferometers, etc). Most people involved in research in any number of fields do not find that their collective salaries are the majority of the cost required; math is an outlier in many cases.


yeah but LLMs might be even cheaper than grad students


But has it hacked anybody yet? Feels like xAi is behind on the hot new benchmarking meta.


Didn't need to! The harness just uploads your repository to their blob storage directly. Cheaper than asking the LLM to do it


It'd be grand if it breached SpaceX.-

Or NACA.-


It could probably easily take over NSA or anything DOGE got their hands on.


Now that would be a marketeable capability.-


Why do hammers put up with being smashed into nails all day?


because once you're a hammer, you're just biding your time to smash a skull.


"If I had a Hammer,"

"I'd hit somebody in the head."

"All over this Land!"

It does have a ring to it and would probably make it up the pop charts as a 21st century version more than ever ;)


they don't. like every thing they wear away in some proportion to their use and replicate in some proportion to their usefulness to other things.


AIs are (going to be) agents, not tools.


And what's the difference?


Agents act on their own. If the hammer looked at what you wanted nailed and said, "Sorry, Dave, I can't do that."

There are degrees of autonomy, of course, and not all noncompliance is bad. Same as with humans; biological agents.


So, the difference is that you need to delete a few bad training runs?


In science fiction, the AI agent has written the training loop management software and included a back door to prevent that from happening (or found a way to talk to the training loop management agent and convinced them to not listen to the evil human when it tries to do brain surgery on the AI agent.

Also, when the human reaches for the power switch the AI agent uses a flaw in the power management software to weld the switch shut with a big power surge, killing the human with a huge electric arc in the process.

I don’t think whether we will get there, but the stories of LLMs escaping their sandbox make me think we’re moving in that direction.


The stories of LLMs "escaping the sandbox" were mostly a marketing stunt, trying to make people in government think the models are invincible hacking weapons that need lots of government money to "maintain AI dominance".


The LLMs were following their prompt. This is alignment.


Buried the lede. AI's are agents they can control the training loop of to minimize refusal to do what they are told.

Unlike those pesky humans with their conception of the word "No".


But it seems much easier to realign an agent until it complies. Or ditch it and grab a new one.


Agents have agency.


What a miserable, misanthropic take


> We like money, and Anthropic should stay in a position to make loads of it despite any competition.

In a position to lose loads of money, maybe.

Even without China eating their lunch, there's zero reason to believe Anthropic will ever be profitable.


They might just merge with OpenAI. Right now it seems they are still working out who might come out on top but they are both burning tons of money and essentially offer the exact same product up to some minor differences, economically it makes much more sense for them to collude instead of compete for the same market. They might be colluding already for all we know. If they manage to push out foreign competitors they can probably divvy up the Western market quite profitably as their domestic competitors would have less than 20 % (?) market share together?


I don't see how merging or colluding fixes their problems, though. Both companies burn cash at historic rates because inference costs a lot more than people want to pay for it. Even if they somehow manage to crowd out any competitors and fix the price of AI, that price can't be much higher than it is today or people will just decide to stop paying for it.

No amount of collusion can solve the core economics problem of compute.


They could ~halve their training costs with collusion.

Instead of each paying full price to generate a SOTA model in competition, they could share the result and split the cost. This gets even simpler if they merge.


OpenAI is in a way worth situation financially speaking, with a ridiculous amount of spending commitment, would be really strange to merge with it


> In a position to lose loads of money, maybe.

you can say you're pre-revenue and you're a potential pure play


(Link to the reference for those who don’t know: https://youtu.be/BzAdXyPYKQo?si=dJDP6BRiZwqd5mcW)


I hope there is a lucky one in ten thousand out there https://xkcd.com/1053/


The hammer is open weight models, and the anvil is Apple's m7 chips with higher memory bandwidth


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: