But at the same time, nothing specific was asked in the prompt, so the boring result may arguably be what is the most aligned with the original request. Personally, I wouldn't want a model to add fuss to something while I never asked for it.
The more I read about everything that has been written regarding AI regulation since the OAI/HF incident, and the more it reminds me of the nuclear arms race (although the potential consequences would possibly be very different). Surely, we managed to negotiate a Non-Proliferation Treaty with most of the countries of the globe. Cannot we take inspiration from that for AI?
I also think the nuclear arms race is a good comparison. I think in hindsight, we've gotten extremely lucky with how the development of nuclear weaponry went, in ways we probably won't with AI.
1) Right after nuclear bombs were first developed, they were used in an extremely public way to kill 200 000 people. This was enough of a cold shower that everyone was agreeable to a global non-proliferation treaty. With AI, we might not get any warning shots that kill 200k people before the incident that kills or disempowers all 8 billion.
2) The fairly useful related technology of nuclear power is nevertheless different enough from plutonium enrichment that countries can have power reactors while provably not doing anything weaponry-related; and also nuclear power is not that useful a technology, so most countries can do without. With AI, we have neither of these factors: the intelligence that makes a model useful is exactly what makes it dangerous, and the economic incentives to do AI research are immense, with pessimistic estimates like "automate a lot of intellectual work" and optimistic estimates like the singularity. That means that if we do get a warning shot where a misaligned model stupidly kills a few million people instead of biding its time, it's not guaranteed that it'd be enough of a cold shower to trigger negotiations.
So negotiating an AI non-proliferation treaty would be much harder than the NPT. I nevertheless hope that at least a few world leaders can understand these considerations and figure something out, because it's probably our only chance. As far as I'm aware most of the technology for letting parties verify what the other parties' compute is used for already exists, it's only a matter of diplomacy.
> Right after nuclear bombs were first developed, they were used in an extremely public way to kill 200 000 people. This was enough of a cold shower that everyone was agreeable to a global non-proliferation treaty.
Not really. Before NPT you missed a small gap in there of 20 years where the USA and USSR built 10's of thousands of nuclear weapons and ICBMs.
And actually, public perception at the time saw Nuclear Weapons extremely favorably, as the bombs dropped on Hiroshima and Nagasaki ended the deadliest war in human history that killed 50-60 million people, most of which died excruciating deaths in trench warfare, fire bombing, chemical warfare, starvation and so on.
I mean, sure, the immediate cause of the NPT was the ability of everyone involved to foresee the possibility of a global nuclear war and judge it both worryingly likely and catastrophic. But I argue that the Hiroshima and Nagasaki bombing was a major reason why this possibility was salient, rather than being treated as baseless conjecture. Whereas right now most people do treat the idea of AI x-risk as baseless conjecture, and this would be different if there was a "warning shot" to point to.
> Surely, we managed to negotiate a Non-Proliferation Treaty with most of the countries of the globe.
N/S Korea, Russia/Ukraine and finally US/Iran changed everything in regards to stances on nuclear weapon ownership for many countries seeing pressure for the great powers.
Every country that can get a nuke will, and if given the opportunity will use LLMs to speed up that process.
It might help if we stopped talking about AI as if it is itself responsible for its actions and excusing the humans who create and operate it. People should be held accountable for the behaviour of their software.
Why am I reading fantastic stories about swarms of agents struggling with moral dilemmas instead of reports on the lawsuits and criminal investigations that would surely ensue were the software involved not called AI?
The big difference between nukes and AI is that only a handful of people in an even smaller handful of countries know how to make them, so coordination is easier to acheive.
Everyone of any level of intelligence can run a frontier-class model if they have the GPUs. It's like trying to coordinate swarms of mosquitos.
nukes arent actually hard to make, its just takes a lot of pretty visible tech a long time to do, so its quote obvious whats happening.
the bigger difference IMO is that frontier models are economically useful, where nukes are basically dumping money into something that doesnt change much in your bargaining power
> This could be seen as analogous to the SALT treaties — capping the number of missiles limited the potential for destruction while preserving each country’s deterrent
That's a nice idea but I hate how every instruction step in the README is just "Ask your agent to do it". Not only does it make it harder and more time-consuming to understand how the general system is implemented, but it also dumbs down users. Do you think people checking your repo wouldn't be able to copy/paste a few commands or find an equivalent on their system?
that's the neat part, there is no prescribed system or code. your agent decides the implementation which you control, the repo is mostly natural language (save for the presentational tui)
The recklessness is moot though. We don't have direct access to how the models were prompted and asked to respond to certain events, so they could well have been incited to commit fraud or other illegal actions, in which case the persons in control should be held accountable as perpetrators.
You would have to prove they acted intentionally, though. You can't just argue in court, "Well, we don't know how they prompted, so we will assume the worst"
If the prosecution is able to prove beyond a reasonable doubt that the person gave a prompt that was intended to commit a crime, then of course we can prosecute them for that. The AI is just a tool to commit fraud at that point, and is no different than a person who uses photoshop to alter a check to commit fraud.
Pragmatically, if too many people use LLMs recklessly, then we ought to regulate them.
Of course, it'd be better to not regulate, keep LLMs users reponsible and publicize this reponsibility in order to mitigate damage. But if this is not enough then we will have to move the needle somehow. Similarly to guns, drugs and so on.
It has nothing to do with whether you know something is a crime or not. Many crimes require intention, such as murder and fraud, so you have to prove that to convict someone of that crime.
That doesn’t mean you can get away with anything as long as you don’t intend to commit a crime; there are many other crimes that don’t require intention at all, like involuntary manslaughter or gross negligence.
My point is that if you want to charge someone with a crime that requires intention, you have to prove that intention.
On the same topic, it's been a while since I have tried to find the best product to treat canker sores, but I have never found anything better than solutions composed of carboxylic/mineral acid and alcohol, and even then, it hardly suppresses the pain and I have found the treatment time to be about the same whether I apply something or not. The dream would be to have something that make them disappear near instantly.
But dental scalers don't offer the advantages water flossers do for gums, do they? Plus, from whet I've read online, it seems that these devices need to be used after brushing, as opposed to water flossers which are used before brushing. And for dislodging food (especially with braces), water flossers remain the best tool.
I use my scaler before brushing and I use it for getting off certain things that bother me. It beats going to the dentist for some chronic issues I may deal with. I would say it should be used in conjunction with a waterpik as it would be far too abrasive/damaging toward the gums, for sure.
Psych 101 says stuff like people being more willing to accept a date with a person after a scary experience since they mistake their quick beating heart for love, rather than humans sees a scary experience as a stronger signal of bonding than a casual experience.
There is so much simple tests that they draw extreme conclusions from in the very first lectures you hear, and when they teach those things as true when its highly debatable its hard to not throw the entire field in the trash.
Of course there is probably some good psychology work done, but you shouldn't have so many shoddy examples in the first course then since that gives people a bad taste for the entire field.
Take physics 101 (or something, anything from a hard science) and understand how they know what they claim to know, and compare it to psych. It's night and day.
One doesn't need to continue learning a topic to know it's bs when it's bs right from the start.
A 101 course should not teach frontiers of research. It should have fundamental established theories upon which the rest of the field is based. My experience with introductory psychology textbooks is that they spend a lot of time discussing ideas discarded decades ago, and then they spend some time discussing ideas that were discarded after the textbook was written. There is typically one poorly explained chapter on neuroscience that the authors summarized from another textbook.
reply