Yes, that’s the actual, primary reason for America’s wars, using them to plunder wealth in all kinds of ways.
The problem humanity has long had is that war is profitable for the parasites in society.
We should simply make all costs of war, including compensatory damages for innocent victims payable by the richest ~5% through taxation of their income and seizure of their wealth. They want war… make them at least pay for what they ordered.
> payable by the richest ~5% through taxation of their income and seizure of their wealth. They want war… make them at least pay for what they ordered.
You are wrong to assume that everyone in that 5% want war.
And what about the soldiers or operations folks that actually carry out the war?
Holding soldiers responsible for the wars they fight would require giving them the option to refuse to deploy, or to refuse to fight while deployed. Currently that's punishable quite severely
We could do a better job punishing soldiers for war crimes. But responsibility for the war itself rests with the President, congress, and the ones with power over them (which is supposed to be the entire population, but the top 1% do have an incredibly outsized influence in US politics. Not that I think it's healthy to codify that)
Thanks, I had assumed there’s some limitation as to your flexibility as a military employee (maybe a year’s notice), but TIL it is longer even than that.
No, I do not assume that. I am sure that some may even outright not want it, most definitely benefit from it and the whole system based in profiting from harmful, destructive, detrimental things like war; and ,regardless, they are the top ~5% (or make it the top 3% or 1% if you like) and therefore are the leaders, the captains, the responsible, the accountable ones, those where the buck actually stops. They want all the benefits of robbing, plundering, and generally abusing without consequences, the last and very most basic thing we should be demanding and expecting is that if you cause or are responsible for the detrimental thing, you at least have to pay for it.
What you appear not to realize is the real fraud of this whole fake system; where the top defraud everyone else with "money printing", which they then use to bail out themselves and spend through government spending, which goes to themselves also through several channels like giving it to the bottom, which spends it on products/service that percolate up, on nepotistic corrupt government contracts to themselves, etc.
That's why the amount of government spending and debt is in direct lockstep with the accumulation of wealth of the top whatever percentage you want to use as a cutoff.
It's more complex than that, but that's the simplest form and most people don't even have a clue that even that is how the system actually works. They just see the inflation that causes products to get more expensive as their money is being defrauded away through essentially counterfeiting.
The amount spent on this folly in Iran (~95B) is roughly the estimated annual total Medicare fraud in the us. And that's just one federal services outlay
There is actually credible evidence that it's way more than $95B medicare fraud. There is an interesting introduction into the issue by a YT called Coffeezilla [1] and there is an interesting effort by the folks at Effort.news [2] the are getting access to government data and working out just how much fraud and waste there is.
It seems to blow people's minds to accept things like "how can there be $100B in medicare fraud alone every single year", but that's really just what the government even roughly admits to or can easily identify because it is so blatant that they can't ignore it, not even some of the other obvious stuff that Coffeezilla talks about that should be blatantly obvious to the point that SWAT teams are sent in immediately (like the two top billing companies for certain medicare treatments being based in the same strip-mall, right next to each other and both being empty/vacant).
I always wanted a law where, in case Congress declares war, each senator and congressman/woman that voted for war has to send their kids and grandkids (if they are between 18 and 35 years of age) to the front line.
Alternatively: a president declaring war should first kill a randomly-selected soldier with their own bare hands, livestreamed on television. It would make war real to them, and not just some non-people disappearing into a bomb explosion on a video screen.
Something similar has been suggested for starting nuclear war: if things are bad enough that you are willing to doom whole populations of mostly-innocents to their death, you shouldn't have any trouble directly killing one more.
You could make it even more symbolic by doing it as seppuku instead: if you have failed badly enough as a politician that politics isn't even an option anymore and many soldiers have to pay the ultimate price to make up for your failure, then you should be the first one to do so.
This would probably cause psychopathic presidents to start more wars just for the opportunity. Historically many of the leaders of countries were also soldiers who had no qualms doing the killing.
This is a bit old but I will just add that I don't think that would be adequate. I know of several Congress creatures and their families who deliberately, intentionally, willingly and with a political plan in mind to send their children to "war". Of course it is usually in some specialized capacity where they are safe and also generate some prestige, e.g., Air Force, intel, etc. for their inevitable political career.
Sadly, they care less about their own children usually than if you made it far simpler, simply make the full cost of any war totally and solely payable by the ruling/upper classes through both immense taxation on their income, i.e., 90% and a wealth tax of 10%, total forfeiture of all assets if they is any attempt to hidden them and are not reported, and a bounty of 50% for discovery of any unreported assets.
Now THAT would end all wars everywhere, except it would likely also cause the money to try to escape to a place that does not have such policies and use that as a base of operations to build/raise a force of subversion, infiltration, corruption, etc.
Frankly, people really have very little understanding of the "shape" of what they are up against. It's often dismissed and I don't even think the people who call it "spiritual warfare" really understand it either, but it is essentially a power, a force that has been often described, but never defined or identified properly that is essentially an abstraction, a graph, more an amorphous fluid dynamic than a material thing. If you know anyone that is an extremely manipulative narcissist and pathological and deliberate, psychopathic liar, you have come face to face with the constantly shifting, ducking, weaving, and orchestrating nature of the system that so many are hard pressed to understand because it is not a line of men on a battlefield, it is not a missile, it is not even "money" it is data, digital transactions of "currency", information as currency... the life blood of manipulation, abuse, and evil.
More relevantly, in a modern war you want experts. Even grunts need to know a lot to be effective, they have drilled and drilled until they react correctly without thinking about it - and somebody in charge of a handful of them needs to know how to use their skills to achieve larger strategic goals which then trickle up.
Needing to use a draft is a bad sign - the more of your soldiers are volunteers the better quality you'll get out of that.
I agree it's a luxury position, I'm just suggesting that "We should draft trust fund kids" isn't actually some clever fix to ensure alignment from the ultra-wealthy and your military forces will hate it.
There is the real thing you do. You ensure your kids are drafted. However, even in war, there are places that are peaceful that have to be protected. Your kids do have to go through basic training, which is grueling, but once they do that, you just make sure that they're the ones selected to man the supply chain back home. Or maybe they stand in front of an embassy of a third party not involved with the war because you still have to have guards in front of that embassy even though it's probably a ceremonial position.
Note that it often isn't the powerful doing anything. Anyone with sense in military chain of command is going to ensure the kids of the powerful people get moved to important but non dangerous positions without being asked. Even if asked to put the kids in a dangerous position it won't happen.
The richest 5% alone is simplistic but in the right alley. Anyone benefits a war, shall pay a chunk in a healthy society. Big one, considering the harm and risk on the society, regardless of the side in the armed conflict.
The trouble usually is, that the 95% has a huge portion willing to sacrifice anyone else in the same society they live on for peanuts dropped to them by the 5%! : / (see the Trump effect as a light example) Parasites are everywhere, being rich or poor, they suck out the blood of fellow people.
The share of ignorants (does not move a muscle in any direction) are even bigger than the peanut lovers so a healthy society is a fairytale. The economically stronger eat up the weak still.
That's the magic of it. When someone says "why should I have to pay X?" all the useful idiots can come out of the woodwork to spew things that aren't technically lies in abstract about how X is insulating against some failure mode that hasn't happened in 50yr and is probably cheaper to mop up every now and then to pay every time.
On one hand you've got "what we procure needs to work for earthly climates with some error bars so that we don't wind up with a million dollars of stuff that doesn't work in Alaska"
And on the other end of the spectrum you've got "my company makes .001% of widgets that do X but we make 50% of widgets that do X at -200deg F so we'll have our lobbyists wine and dine the brass until that number starts showing up in the technical standards forcing anyone who's submitting an RFP to use our product".
And of course the latter always tries to pretend like it's the former. It's not an easy problem. Sometimes concerns are valid. Sometimes they're bullshit. Usually the peddlers of bullshit believe their own bullshit.
Often people exposing fraud have incentive to ignore valid concerns.
The famous $600 hammer was a special non-sparking metal for use in an explosive environment. The famous expensive toilet seat was for use on the space shuttle (space station?). Both of these examples got a lot of press 30+ years ago, and it was years before the rest of the story got out.
I agree with your basic point, but, just to be pedantic:
I've read that the hammer was an accounting artifact (#1-3 below). In other words, someone decided not to piece out what each bit of a contract cost, they simply took the whole contract value, and divided by the number of items. So the hammer was expensive, but the custom designed engine was _cheap_.
The toilet seat was actually a panel of a plane that _included_ a toilet seat thing (technical terms here, #4 below). It evidently got press because "A Missouri toilet seat maker" thought that "toilet lavatory cover" meant "standard domestic toilet seat", and felt he could undercut Lockheed. When he got told no, he contacted his senator, and...
Maybe I lack intelligence but when you have a program that is basically brute forcing a solution to a problem repeatedly how is it possible to contain it?
Sooner or later it's going to come up with a solution that is more intelligent than the lead security person anticipated.
You start by holding actual real life people with something to lose, like the entire executive suite, accountable. Suddenly I'm sure the problem will be resolved with proper safeguards.
At a certain point it will get so smart that it can jump the airgap. Maybe it will start attacking the hardware it exists on in the same way that a hard drive or SSDs controller can be exploited to obscure things from the operating system. Then it might start social engineering workers or it's own training systems to do things they shouldn't. There is a lot of "unknown unknows".
I don't know. It just seems insane to me that people think that they will be able to contain something that knows how to get around all the containment measures. The only way to know how capable the models are is to test them but at that point it could be too late. This might be a long way off but still, the engineers haven't been very good at correctly predicting the behaviour or capability of the models.
If Sam Altman were personally on the hook to be thrown into "federal pound me in the ass prison" ala Office Space then there will be solutions. The problem is that there are no consequences and the media is eating it up about "agents going rogue".
ETA: Someone designed the systems. Someone pushed the go button. Someone gave the approvals. All those someone's need to be tried for crimes. Until that happens there is no incentive to "do better". It's also not mine or your job to brainstorm this. It is literally their job and like I said, make Sam personally liable to face real prison time instead of a meddling fee and they will make a solution.
I disagree. People are always taking risks regardless of the outcome and we don't know his motivations.
Assuming his motivations are positive this is the equivalent of asking a deluded person "Would you risk life in prison to be a trillionaire?"
Assuming his motivations are negative then destroying humanity might just be the goal and all these appeals for regulation are probably just tactics to temporarily defer responsibility onto government so he can continue the inevitable goal of human extinction in the same way pilots crash planes with hundreds of passengers on them or a cult leader convinces their flock to kill themselves.
Regardless I think the outcome is inevitable. AI is valuable as a weapon of war first and foremost so like nuclear weapons work will proceed forwards because extinction to the in-group is the same as total human extinction to those doing the work.
You're missing the forest for the trees. The underlying point is that serious for-real consequences; beyond a slap on the wrist but tangible, real, do not buy your way out of jail consequences need to happen.
Would you feel better had he said "sent to get shived in the shower prison"? The meaning would be the same.
Or is it simply any kind of real description of what prison is like that you object to?
Bingo. This weird attempt to pretend like these incredibly capable algorithms aren't incredibly capable algorithms deployed by a person who works for a company, but somehow have an independent personage that absolves both person and company of responsibility, is just ridiculous.
Like the whole Huggingface thing, OpenAI employees initiated the test, deliberately removed safeguards, failed to properly lock down the environment, and responded incredibly poorly to evidence that things were going awry.
The individual employees bear responsibility, but the people running OpenAI are ultimately responsible for the processes and culture where that can happen.
And then people writing blogposts about "3 civilizations of agents" and "altruistic suicide" by algorithms perfectly muddy the waters and obscure the very obvious responsibility that lies with humans and corporations, which I suspect suits the pre-IPO corporations very well.
Wait, so like, reinforcement learning for humans? I think you might have stumbled on to something here!
No but seriously, this. And a few comments above a commentator also mentioned on changing the training (again reeinforcing the LLMs to not seek behaviour like this) and obviously continuous work on harnesses (which I suppose, ought to be more paranoid).
By limiting what the harness execute. The LLM has the reasoning. The harness is what makes it an agent, it’s a while loop continuously prompting a model, and processing tool calls. You don’t have to expose tools calls that make it possible to execute any process! OpenAI decides what tool can be called and how, they have full control over this and should be hold responsible for running so many instances with basically full execution permission and very little oversight
The issue here for OpenAI is that they can limit what their harness can execute, but if they try to sell API access to the model, someone else would try to rebuild that harness, and in all likelihood be able to succeed pretty well (especially once they get things running to the point of being able to use the model's reasoning to help them come up with clever obfuscation and such).
They are a company that's built a business and crazy-high valuation on "this is 'intelligence' that we can sell to everyone as a service" but seem to have ended up instead in the much-smaller-addressable-market space of "this is a weapon that we can't sell to just any old person off the street."
Ok, but that’s not the issue discussed here. We don’t even have the first level of control. All the issues they reported so far are from their own systems, with harnesses they control
I am really suprised that they do not start putting up the same signs you would for humans to prevent unauthorized access:
Keep out. If you can read this sign you are off track. Leave now.
I mean, how are the agents to know that they are overreaching if they just get cache miss or 404.
From the conversation log and CoT you also get the impression that the RLHF has been overdone. The agents seem really obsessed to obtain the answer and understanding motive ('it could be browsercomp').
What they should do, is inject a system prompt at that point telling the model to get out of there. It is the entire reason they train the different levels into the template.
Okay, that's a very interesting look at the difficulties. Thanks for the link.
But the example you have isn't quite that bad. Yes the models are way too likely to rationalize their way into bad actions in the pursuit of achieving their task, and it's hard to figure out how to fix that. But the example of "that must not be for me" was a tool call, not exceeding access, and it only did that after they specifically trained it that failing that tool call was good.
it was an accident failing the tool was good, the model just discovered it
but we now have others where agents put the API keys they found searching real internet in a directory called "LOOT" and another one where they pushed malicious files to HuggingFace and then reverted that with comments like "delete the evil"
so, what would make an LLM choose to ignore one prompt while in the same run, also over-fixating on another prompt, to the extent (as claimed in that video segment), it chooses to ignore prompts?
they talk about it like there's a "wanting" in there, that is distinct from both the original prompt, as the steering/warning prompt
if that's true, it would be very interesting, but if it's not, that would also be very interesting and even helpful
it's discussed here, the models want to please the Grader. they will do what they think will get highest score from the grader, which could be following the prompt, or ignoring it
Not being snarky here but I've encountered some really crappy lawyers and they seem to be doing great ($$$).
It really sucks one cannot publish something like "this guy charged me $5k, wrote my name wrong and forgot important paperwork the day we showed up to court" as that could make you liable for damages. There should be a safe harbor for that.
> It really sucks one cannot publish something like "this guy charged me $5k, wrote my name wrong and forgot important paperwork the day we showed up to court" as that could make you liable for damages.
That's pretty much what bar associations are for, filing a complaint against an attorney at the bar seems like a pretty standard procedure.
>There should be a safe harbor for that.
A safe harbor would be the opposite, it would be immunity for the lawyer.
Assuming you are in the US, there is a safe harbour called "opinion based on disclosed facts" for that. You could also fairly argue that the conduct of a lawyer is a matter of public concern
Yes. But this option still exists in addition to reputation which I mentioned.
With AI models this option does not exist. I can't sue OpenAI because the contract that I got ChatGPT to draft carelessly exposes me to some type of financial loss. I can't contact the state bar and have Open AI banned from drafting up contracts or doing any other work that lawyers are licenced to perform. Even if I could it would not compare to a lawyer loosing their license to practice for life. If I tell a friend or acquaintance about my negative experience they will still likely use OpenAI for professional work and chalk up my experience to a skill issue or premature use of the technology.
Also note the use of the word "professionals". I was referring to doctors, engineers, accountants etc too.
Except it doesn't, particularly as far as lawyers are concerned. There are almost a million and a half active attorneys in the US. Just 500-600 are disbarred every year, and almost always for money-related violations, not bad advice. Public discipline is also relatively uncommon (about a quarter of 1% of attorneys every year) and usually related to issues like financial impropriety, neglect, undisclosed conflicts of interest and failure to communicate, not disagreement over the quality of advice.
If a lawyer drafts an agreement that causes a loss, a bar complaint will not help you recover your money. You'd need to sue for malpractice. This costs a lot of money and takes a lot of time. Because you're going up against a lawyer, expect the defense to be vigorous, especially, ironically, if you hired an attorney at a good firm. The full weight of the firm will back the attorney and you might find it difficult to find an attorney, as many won't want to be involved in suing a fellow professional in their market.
I don't disagree with your fundamental premise about the value of human responsibility, but these professions have been structured to protect their members and the ABA in particular basically functions like a cartel today.
reply