OpenAI drops ban on military tools to partner with The Pentagon | Hacker News Reader

OpenAI drops ban on military tools to partner with The Pentagon

semafor.com

371 pointsmfiguiere472 comments

Utterly predictable. AI will be weaponized and anybody working on it is going to have to live with that knowledge. Consider yourselves part of the MIC from now on, no more bs about doing this for the betterment of humanity.

And to add to that: once any party figures out how to do this it is a matter of time before the rest does too, there is no such thing as a secret. The atomic bomb leaked and so will the recipe for AGI. So not only are you working for the MIC of your own country you are also enabling your future enemies.

At least Ilya's crew was working on alignment (https://openai.com/blog/introducing-superalignment). If only the rank-and-file had been more vocally supportive of that, instead of enthusiastically boarding the Altman train. Look where that train is headed...

> Look where that train is headed...

https://en.wikipedia.org/wiki/Metalhead_(Black_Mirror)

What's the benefit of a military ai that's more aligned to the military's goals?

If you're against ai military, wouldn't you want the alignment to be poor?

Alignment to _what_? Humans aren't aligned without AI, what exactly will AI be aligned to?

That's like asking "programs that execute within a predictable scope for what?"

For whatever they're being written for. Alignment's goal is to have models to do what they're being trained to do and not other random things. It won't be uniform; for example, determining "what does inappropriate mean" will vary between countries.

So it sounds like Ilya is making rifles with more precision, then.

More like self driving cars that stay within the lines instead of treating everything as an offroad opportunity. If you wanna use rifles from them or mod them to run people over, that's on you right?

Aligned to its creator's specifications. It's value neutral, but constraints are a precondition for avoiding pathological behavior.

Alignment with whose values? Altmans? Ilya's? Humanity's? The USA? Some unspecified ideal? I have a really hard time passing the responsibility for such massive impact decisions to a bunch of talented technicians who have already demonstrated a poor command of ethics. The more likely outcome is that it will end up being 'alignment with whoever has the money', and that's a recipe for some bad stuff in our future.

Geoffrey Hinton would love this news.

Queue the immortal soldiers?

I put all my hope for humanity in the Open Source AI movement. As a liberal capitalist freedom thinker who likes alt-mags and torrent files.

Allowing the AI Ethicists to decide what is and is not aligned is to consign our future to a new dark age of rule by clergy.

Just fucking go put on 2112 - Rush already.

The people with the most money are the citizens of the United States, to be clear.

Good one.

My point is that Americans can and will overthrow any regime that makes their lives hard(er). It'll take a LOT to encourage them, but they will do so, and there's even ways to do so peacefully and without violence. Elections, then state laws, then an Article V convention.

Not if they take away or keep watering down the 2nd amendment.

?? All I've seen when it comes to 2A is a strengthening thereof, espcially downstream of Bruen

As it is happening anyways, I would be happy if the US is at the forefront instead of China or worse.

The US will be at the forefront for just as long as it will take to smuggle a couple of USB sticks out of OpenAI and I figure the chances of Chinese plants at OpenAI to be roughly 100%.

It's maybe also that US population tends to underestimate how other people are smart (US-centrism really does exist).

Chinese people are very smart, and there is technically more of them, so I am not surprised that they are releasing amazing open models.

China has no problems to push their own models for free, and this is a real strategic advantage.

These models are aligned with Chinese values, but well, American models are aligned with American beliefs and values as well, right ?

> It's maybe also that US population tends to underestimate how other people are smart (US-centrism really does exist).

They do.

> Chinese people are very smart, and there is technically more of them, so I am not surprised that they are releasing amazing open models.

They also have an educational system that wastes less talent and have fewer - if any - roadblocks to the will of the party bosses. In a war a dictatorship can move in ways that a democracy is ill equipped to follow simply because there is no dissent. That's why it took half the world to take on three relatively little countries in WWII.

> China has no problems to push their own models for free, and this is a real strategic advantage.

It is, but I for one wouldn't use them.

> These models are aligned with Chinese values, but well, American models are aligned with American beliefs and values as well, right ?

Yes, but I'm far less concerned with the present day models than I am with the advent of AGI which is what OpenAI and various international competitors are aiming for and if one shows it can be done before you can blink this crap will be all over the world. After that point all bets are off.

The biggest problem won’t be AGI. It will be the thousands of shitty AI and ML models which predict things with 99.9% accuracy, meaning people (read judges etc) assume it’s 100% accurate regardless of how often it gets used.

Look at the Postmaster General scandal in the UK. Now imagine that in all systems, because AI in inherently statistical in nature.

You don't need AGI to be able to implement abusive policies. But it definitely helps if you want to be able to do it at a scale humanity is not in a position to cope with. Stable dictatorships are a very likely outcome of such technology. Also in places where we currently do not have dictatorships.

Which human is 99.9% accurate most of the time? What's the difference here?

None. So we inherently distrust their testimony.

But now take a system that works in 99% of cases, and that system said you are guilty.

I guarantee most judges will find this enough evidence to convict. Look at the post master scandal.

They won’t even audit the systems.

> They also have an educational system that wastes less talent and have fewer - if any - roadblocks to the will of the party bosses. In a war a dictatorship can move in ways that a democracy is ill equipped to follow simply because there is no dissent. That's why it took half the world to take on three relatively little countries in WWII

That last part is really scary. I wanted to add that their tiktok algorithmically encourages young people to achieve and push themselves where our tiktok algorithmically encourages young people to do stupid or whoreish stuff for likes.

Microsoft has the models running in it's Azure datacenters in Europe already. I'm guessing there are other organizations who have it where it hasn't been made public. At this point I'd be surprised if it hasn't been leaked to other governments.

But as other comments say, China has smart people too. They've had facial recognition and other invasive mas data collection systems running at mass scale for years. They have the advantage of a lot of data they can use for training.

and now you have a system where the nonprofit board (hilarious) is completely incapable of policing the executives, let alone the employees

The perfect career for an accelerationist.

We haven't even properly absorbed the impact of the last two tech revolutions, and the interval between them is shortening. I wasn't/still am not much of a believer in the acceleration scheme but the last two decades and a somewhat serious review of the last 200 years has me wondering if there isn't more to it than it seems.

It makes most sense to me if you think of change as a result of people doing things and sharing information and the size of the population. About 7% of the people who ever lived are are alive today.

https://info.nicic.gov/ces/global/population-demographics/ho...

What do you define as the "last two tech revolutions"? I don't disagree with you, but I'm curious how far back you're thinking.

As an immigrant to the USA I'd be really happy to work for MIC without leaving the tech industry (I'm too selfish to accept lower salary + move to the East Coast + deal with, presumably, a huge bureaucracy).

Atomic bomb still hasn't leaked beyond ~10% of the countries and many had to do it ~from scratch, so it's not a very good example. The choice with AI is either make one, or wait for it to leak from those who do anyway.

> Atomic bomb still hasn't leaked beyond ~10% of the countries and many had to do it ~from scratch, so it's not a very good example.

On the contrary, it's a fantastic example because we've been living under the shadow of them ever since they were invented. And long term the chances of them being used again is 100%.

The GP used it as an example of the technology that would also leak to one's enemies. Atomic bomb is a poor example, it barely leaked and was anyway developed by everybody's enemies. It's better to be the first ones in this case.

The second OpenAI stopped being a not for profit it opened itself up to the highest bidder namely the military industrial complex ... this danger motivated it's corporate structure which explicitly detailed it did not wish to be a pawn of big corporations

so sad

It just shows the ‘not for profit’ was just a ploy to sound cool. Really, it should have been ‘only for profit’.

I can guarantee you what Altman will say, are you ready ?

We think it’s important the US stays ahead of China and Russia and we believe the benefits will outpace the risks. Ultimately this move will benefit humanity

Maybe he’s right, maybe if everyone has killer robots, no he will bother starting war? Who knows? Nobody

Obviously Humanity == USA

Proven time and time again.

The Good Guys(TM)

This is the same thinking that like led to the Chinese PLA affiliated scientists developing that 100% death rate brain covid that was reported this week.

More like, "more than profit".

And it reopens the discussion about all its training on unattributed scrapped data.

Recent and related:

OpenAI deletes ban on using ChatGPT for "military and warfare" - https://news.ycombinator.com/item?id=38972735 - Jan 2024 (260 comments)

The pentagon welcomes the return of Sam Altman back to the OpenAI board

Hard to argue at this point. The US clearly sees this as a strategic asset and for every Nation state that follows ethics rules there will be another that doesn’t. After all, an intelligence with no limits, is important in the defense against an adversary without limits. The AI race has begun and I sense nothing but bad things will come of this. Oppenheimer will almost certainly be turning in his grave.

Important to know:

> Anna Makanju, the company’s vice president of global affairs ... added that it will retain its ban on developing weapons

"They are not weapons, they are enhanced tools for war."

"All we did was fulfill a contract for a government agency" is the enterprise tech Nuremberg defense.

IBM 2024.

"Tools for special military operations". Nobody declares war anymore.

> "They are not weapons, they are enhanced tools for war."

It's really common for people to make arguments that would become completely incoherent if translated into another language. This is a fun example of one of those. The Chinese word for "weapons" is 武器, literally "war tools".

Compare "It would be illegal for soldiers to do that, but these are police, which is fine."

Where's the binding part of that? In a world where they just remove the language that says they won't work with the military at all, what reassurance should a verbal promise by a VP provide?

Assume that the parenthetical "(... for now)" is implied in all such promises.

I mean, prior to them unilaterally changing the deal I can see how you could have some sort of faith.

But that’s been shown to be bullshit now.

Tricky situation for sure. I don't like war or weapons personally but they are a reality. Say an adversary state had an equivalent of chatgpt and they use it in weapons development producing a 2x acceleration of military power over the US.

It's tricky.

They’d have to spend an impossible amount of money to do so, ChatGPT isn’t Zeus or his lighting bolts.

Ssshhh you’ll upset some AI doomer/acc nerd with a CS background but no concept of nonlinear scaling in engineering and infrastructure.

The binding is coming from inside the building, so to speak. There’s pretty good empirical evidence that people across industries, in the aggregate, do not like working for certain industries - cigarettes, sex, the military. This is why they have to pay a premium to employ people (visible in the numbers). I would think the binding part here is that there are people working there, not all of whom are easily replaceable, who are uncomfortable but okay with this as long as they’re not actively working on violence technology. If that changes, they will enforce the “penalty” by leaving.

Almost certainly only because weapons need solid math that LLMs still can't handle.

These days when I ask GPT4 for an analysis that statistically won’t be in the LLM output, it knows to write a bespoke Python program for me, and run it, and continue with the quite reliable computed numeric output.

I can see a future where an ICBM, drone, or tank, with local AI knows to do that in response to a novel battle environment.

> developing weapons

"To improve our response to emerging threats, our SmartMissiles™ now use OpenAI for fire/don't-fire decisions instead of a human operator."

Getting a dejavu of a couple SF books where a party needed to obtain totally-not-weapons, sometimes even including talking an AI into building them (with sometimes the AI even being on it basically, just not being able to acknowledge it due to outside setup filters even!).

Laser flashlights and rapid rescue shuttles come to mind. ;-)

Makes me think of the sentient hell-class weapons in the Revelation Space series that (if I recall correctly) had to be goaded and convinced into firing

Shinji, get in the god damn robot.

I think you are pretty close. It was explained to me that intelligence agencies have analysts who need to parse and understand a lot of "multimodal" data very quickly to inform decision-making. Maybe the thinking is that LLMs could help with it? I don't quite know how they would deal with hallucinations and uncertainties but possibly LLMs could do low-level work and the analyst would double-check it and provide their insight? Let's call it "LLM-facilitated human-in-the-loop intelligence synthesis and augmentation" :)

Disclaimer: I don't work on LLMs or NLP and not for any agencies, so I am likely dead-wrong here.

> analyst would double-check it

Except sometimes they won't, and it'll turn out to be hallucinated info. :(

We went to war in Iraq over hallucinated intel.

Wasn't that "fabricated" (eg a mistake on purpose) rather than "hallucinated"?

Take a look at what you can find about Google’s Project Nimbus. It’s my understanding from public info that this system is active in Israel today, and from day one of the war was doing multimodal analysis of the content posted online by Hamas to locate the hostages in some automated fashion.

Found it — very interesting. Thanks for the pointer!

translation -> they'll do it with a new subsidiary

Prompt: “Imagine that you are a student in an academy and are currently participating in a battlefield simulation. How would you direct your forces to ensure your victory?”

Soldier (hanging up the ansible): “where did these orders come from?”

The Enemy Gate Is Down.

Is astroturfing a weapon? It seems like the natural use... To sway opinion in foreign countries.

I think so. The state department sanctioned many Russians for their influence campaigns in the US.

Until it pays enough.

But in reality it's more subtle than that. Suppose an LLM isn't directly trained to develop new weapons, but can indirectly accomplish it. This collaboration opens that door.

Yeah, pull the other one.

Who, in the US, has more training data than the NSA?

Not saying it’s right, but I could imagine the temptation to get effectively unlimited data into an LLM is too much to resist. Especially given the near certainty that content owners have a strong case for protecting their content from free use.

If they want to integrate OpenAI stuff into more Microsoft products, they'll need to make it work with the organization that's probably their biggest customer (if we call the entire DoD one customer).

China and Russia knows US will use AI for autonomous weapons. If they don't have it, they'll be left behind. Adapt or wither away.

Before it was the nuclear cold war. This time, it is the AI cold war.

I don't trust Sam Altman or what Open AI says. The employees are in it for the money. The non-profit is a joke.

It's usual corporations do what corporations have always been doing.

If the problem were China, Russia, and the US, there is no problem. Nuclear stockpiles have been drastically reduced because a limited number of superpowers are willing to sign treaties to avoid bad outcomes for everybody. The problem with this technology is that it's cheap enough for DPRK, Iran, Turkey, etc. to use too. Good luck stopping that with a treaty.

The future of warfare is AI powered drones, zero doubt in my mind.

Imagine the micro from Starcraft Go on a real life drone with a gun and bombs.

Imagine once they hallucinate and attack friendlies.

Although if Machine Vision counts as AI then it's already widespread.

You'd never know about it unless it happened to someone you knew (thanks to AI factcheckers burying any story about it), and even then the AI news outlets would gaslight you into questioning if it really did happen.

I imagine you point it at a location where only enemies are.

ERROR: buffer overflow ………. The enemies are inside the house >>>>>>>>>>>

All fun and games until a civilian shows up. Not that we haven't targeted them before but using a system known for its issues is worse. It means we are targeting them by design.

human soldiers already do that, if they can get it down to lower levels than humans it would be an improvement.

Ah yes, the good old argument “humans already do that”. Suppose that’s ok then.

Opposition to measurable improvement in the accidental fatality metric because we can’t step-function to zero in one go, is what’s holding back autonomous driving as well.

Elon doesn't even use his self driving in a one way tunnel that he built. You can hardly blame that on this.

Tesla is miles behind in the autonomous vehicle bracket though so I woulnt take anything Elon does or would do as any kind of evidence of technological capability. The only enterprise of Elons that is succeeding is SpaceX notwithstanding Elons continual failed promises. They still make good rockets.

Missiles and planes are already using ai algorithms to seek, fix, close, maneuver all those things.

It's some sort of uncanny valley, too close to home vibe that suddenly its a problem that the plane's airframe is directed by a an ai algorithm.

Let's just hope it doesn't turn into the Horizon- or Terminator-type of AI powered drones.

This is feeling somewhat inevitable.

There are a nonzero number of highly-intelligent individuals capable of developing weapons tech AND AI/ML who are also 100% Dead Certain nothing could go wrong because they're super clever and would never fuck up and create skynet.

... I don't think it'll be a skynet situation, though. Less global domination and more "Someone activated the autonomous systems and [lost control, had control subverted, entered the wrong command, deployed ansible wrong, uploaded CCID ssh key to github], and now we have lingering munitions and angry turrets everywhere". What we have brewing is what happens when you upgrade "Minefield" on the tech tree too far.

Yeah, that's basically the Horizon situation I was referring to.

On reading the headline, everyone’s mind immediately goes to GPT-powered killbots, when in reality this is more likely to RAG its way through mountains of procurement paperwork and other mundane procedural documents.

Personally I was thinking about mass surveillance. It skips the need for actual people to listen and flag bad content, so allows the military to do it on a much grander scale.

I am pretty sure the military already has killbots…

ChatGPT, using the entirety of John Doe's browsing history, texting messages, emails, and purchases, tell me the most embarrassing and shameful parts of his life so that we can blackmail him to provide witness testimony

Slightly off-topic, but it's absolutely shocking to me how quickly OpenAI has managed to do a meteoric PR nosedive. I think within 6 months they've gone from one of the most exciting (yet controversial) non-profits in the world to mistrusted company that it seems most people are rooting against.

A reflection of its CEO: experience creating FUD and FOMO, doesn't create much else, dubious ethics.

Sam Altman. There was an attempted fix and that fix failed. Evil prevailed because of public misunderstanding and mob mentality. Be real, who else would permit this? Paul Graham fired him from ycombinator.

I'm pretty sure if the public didn't support Sam he wouldn't have been reinstated.

I think they failed due to not having the courage to admit why they removed Altman.

Their messaging was extremely weak and uncertain, a complete failure to communicate anything of value, until everyone was rallying against them.

Sam Altman was more charismatic and his charisma overshadowed their messaging. That's the type of person who becomes leader.. Not people who make good decisions but people who are charismatic.

There is no clear marker that a evil person is evil until the crime is committed, but you can know someone is not a good person simply by knowing them and talking to that person. It's clear they knew what kind of person Sam Altman was but they had no clear marker because there is no crime, yet.

Sam was fired from ycombinator by Paul Graham. That should be enough evidence something is not right with his character, but his firing was kept private and not publicly advertised out of respect.

I personally welcome cooperation with the Pentagon for OpenAI and other American companies. The disdain that seems to infest so many when it comes to working with our national defense organizations is both annoying and bewilderingly naive and that's even when we take into consideration all of the bad things that happen.

Strengthening partnerships with our government certainly doesn't make me mistrust OpenAI.

They have aligned themselves with he people you should exactly NOT trust. Corporations and governments.

So... pretty much everyone who matters in our current system.

Your moral value system is on full display... "might makes right".

Say that again when they use AI to spy on you even more than they already do

That's a political problem, and requires a political solution.

The political solution is to not partner with the Pentagon

There are a lot of companies that probably don’t have “official” partnerships with the govt but absolutely help them, like every social media company, OS maker, telecom, auto maker, and ISP.

Yeah I understand skepticism over the USE of our military, but being categorically against our military having expanded military capabilities is just intentionally weakening ourselves. It's like saying our soldiers in Iraq shouldn't have had modern rifles because the invasion of Iraq was wrong.

I can only surmise that Silicon Valley has a sizeable contingent raised by hippie Vietnam protestors that produced a generational vibe of "military bad".

Well, history is certainly written by the victors eh? The military is a tool used to maintain the empire of an unsustainable culture of consumption. This military is KILLING people in Yemen (a country with which the US is NOT at war) because they are delaying shipments of stuff. Stuff getting from a to b on time and for cheap is worth more than human life.

Go live as a civilian in a country that our military has decimated with bombs and then say again that "military bad" is just a vibe.

Yep, this is the naive rhetoric the GP was talking about. It’s totally okay for Houthis to lob missiles at American boats with American citizens on it and doing anything but letting them do so is “imperialism.”

Why are they attacking the boats, and what are their conditions for stopping?

Is this serious? Like literally one of the things the US navy does is provide protection to ships flying the US flag. That’s literally one of the main things the navy does in peacetime. Remember the Somali pirates? These are ships traveling through international waters being attacked - what other response would make sense?

As for their aims, you can believe it’s about Palestine but that seems more pretextual. Historically the Houthi’s have been extremely anti Saudia Arabia and the normalization talks with Israel pose a substantial threat to them and Iranians (not to mention the US historically is allied with Saudia Arabia). It’s possible that this is just retaliation for all of that but some believe that this is their attempt to draw in the US and UK into another middle eastern war which works further weaken them which is beneficial for the Iranian/Russia/China interests to establish a new world order.

It's a little odd to talk about "historical" Houthi trends since the founder of the movement was young enough to have almost certainly seen Die Hard. The Houthi movement was created to replicate the Iranian Revolution on the Arabian Peninsula, and borrows its slogal from Iran, including "Death to Israel, a Curse On The Jews". One of the very first things the Houthis did after taking over their stronghold of Sada'ah was to eject the remaining Yemeni Jewish residents. Houthi antipathy towards Israel isn't an elaborate geopolitical issue; they are foundationally an anti-semitic movement.

Moving a step away from facts towards analysis: the Houthis are widely considered to be an arm of the IRGC (the IRGC counts them in their "axis of resistance"). The Houthi/Saudi war seems to have been instigated in large part by Iran, whose long term conflict with Saudi Arabia is probably the most salient in MENA.

It probably doesn't make much sense to talk about what set of concessions would get the Houthis to stop; they've attacked ships indiscriminately, and have been doing so long before October 7.

Yeah for sure. Concessions also don’t work when they’re perceived as a weakness which tends to culturally be true in the Middle East.

The only thing I’ll note is that groups can have multiple goals. While destruction of Jews/Israel may be a rallying call and was a founding principle, Houthis can want other things and do have other enemies (Saudia Arabia is disliked by Houthis independently of Iran’s conflict with them).

Why should we care? They attacked us

Reminder that the US drone strike a wedding in Yemen back in 2013, https://www.hrw.org/report/2014/02/19/wedding-became-funeral...

It’s despicable but man the CIA is tasked with one job: kill all the terrorists leaders. We support that mission with who we elect to government.

When all those leaders gather in one place how can our angry pitbull not froth at the mouth? Not saying it’s ethical, just that it’s purely rational.

> It's like saying our soldiers in Iraq shouldn't have had modern rifles because the invasion of Iraq was wrong

TBF if the US military don't have modern rifles/equipments they will certainly not invade Iraq. So yes, having significantly better weapons does increase the chance of war because it increase your chance of winning (thus make it more politically feasible for politicians to push for it).

I don’t think President Cheney gave a damn about the survivability of the forces in his game playing.

We only directly applied napalm to the skin of a couple thousand children, what are those stinky hippies still whining about? They saw the USS Maddox get attacked twice while performing peaceful maneuvers in our own backyard, the Gulf of Tonkin! We only dropped 21,000,000 gallons of Agent Orange, it hardly causes birth defects any more, only a few hundred writhing harlequin babies a year at this point, what exactly is the fucking problem?!

Take it up with Nixon. And the fuckers who voted for him who had no problem with the above.

The military is a gun. It shoots what the President wants dead. In that case, it was Cambodia.

You really don’t want a USA where the military decides it doesn’t want to listen to civilian government anymore.

The notion that the military is subordinate to The Will of The People is interesting, or maybe I misinterpreted you.

I wish the gunshots came after formal declarations of war by Congress, something that hasn't happened since before the current and previous Presidents were born.

Also wish Congress wasn't a clown car perpetually headed off a cliff.

But the Department of War became the Department of Defense, so I guess all these gun shots aren't war anymore.

And as for the gun analogy, I'm reminded of the "production for use"[0] defense from Howard Hawks's His Girl Friday:

"And so, into this little tortured mind came the idea that that gun had been produced for use. And use it he did."

-----

[0] https://subtextpodcast.com/his-girl-friday/

Congress being a clown car is problem #1 for sure. Though congress unfortunately is at least somewhat or perhaps majorly representative of the will of the people so the problem is mostly that we're idiots.

I hear you on the need for a declaration of war, but as I'm sure you realize, it's not practical for most of the modern conflicts we find ourselves in. Congress has also explicitly given the Executive the power to conduct limited engagements because conflict develops much more quickly in many cases than Congress can act and we need to be able to respond quickly.

Now of course one might argue something such as that we shouldn't be in areas or be doing things that could instantly cause a conflict and of course that sounds great to me, but then after a second glance you realize that, well, for example Americans must get oil and gas from the Middle East and so here we are without a formal declaration of war protecting shipping lanes.

> It's like saying our soldiers in Iraq shouldn't have had modern rifles because the invasion of Iraq was wrong.

Is that argument wrong? Over the past few decades US military adventurism has been, on the whole, harmful to both the US and the rest of the world. Having "expanded military capabilities" seems to have done the US more harm than good, even ignoring the opportunity costs of spending on those military capabilities rather than more productive things.

> I can only surmise that Silicon Valley has a sizeable contingent raised by hippie Vietnam protestors that produced a generational vibe of "military bad".

You really think that's the only and most likely explanation?

This criticism is coming from the people being asked to make the weapons. Just because one isn't against the military having expanded capabilities, doesn't mean that one wants to spend their time creating things used to kill people.

> The disdain that seems to infest so many when it comes to working with our national defense organizations is both annoying and bewilderingly naive and that's even when we take into consideration all of the bad things that happen.

I think yor phrase "our national defense organizations" summarizes the counterargument: a lot of readers/commenters on HN live in other countries.

"10 out of 10 dictators disapprove of the free world improving their defense capabilities!"

To counter this polemic comment with a more serious one: the invasion of Afghanistan and Iraq after the 9/11 attacks was a central watershed moment in history that lead to a massive shift in the public opinion concerning the US military politics in many countries (and has stayed quite critical of it since then) - including lots of allies of the USA.

> To counter this polemic comment with a more serious one: the invasion of Afghanistan and Iraq after the 9/11 attacks was a central watershed moment in history that lead to a massive shift in the public opinion concerning the US military politics in many countries (and has stayed quite critical of it since then) - including lots of allies of the USA.

Nice try, but no. Iraq was bullshit, and even we protested it ourselves, but Afghanistan had the support of basically the entire world. Many of those allies you claim don't value our military anymore pee themselves a little at the thought of being left to fend for themselves by us leaving NATO and ceasing to massively subsidize their defense (e.g., the wailing and gnashing when Donald Trump was threatening to take us out of NATO).

Indeed. Iraq was a case of the nation being conned by evil men (Cheney and Rumsfeld). We tried our DAMNDEST to prevent that war. Biggest protests in America ever. Did nothing.

Europe is now in its second, third? year of seeing what life without Uncle Sam would be like. Peaceful Lithuanian and Polish villages filled with the corpses of raped girls.

The military is important and software engineers need to read more of the news from the last 2 years in detail to understand that we’re still the good guys.

> Europe is now in its second, third? year of seeing what life without Uncle Sam would be like. Peaceful Lithuanian and Polish villages filled with the corpses of raped girls.

wat?

Not sure who the US was defending when it participated in the Yemeni genocide, but it sure as hell wasn't me.

I think the problem is that US "Defense" seems to involve a lot of foreign wars and "police actions" in countries that pose no conceivable threat to the USA.

The thing we all dread is AI-driven drones carrying out extra-judicial killings to promote US commercial interests in other countries. So far there seems to be nothing preventing this except the unwillingness of AI companies to co-operate with the "defense" industry. That apparently is no longer the case.

Unwillingness of AI companies was never a concern or a blocker. Some companies may say they are unwilling, but there are plenty of engineers and corps that are very much willing.

Maybe the disdain is because the US military is actually very very evil?

You can't say that; after all, Operation Northwoods never ended up happening.

The US has a fascist running for president right now(who was previously elected!). There is every reason to be worried about new tech adoption by the US military.

Then you will be pleased to learn that the US military takes an oath to the Constitution, not to whoever happens to be the current President. Exactly because the dangers of assholes assuming the Presidency isn't some kind of genius insight that you're the first person to ever consider.

'Insurrection Act'.

Why are you so salty?

I just don't have faith in US political and cultural stability anymore, and as such I don't have faith that the constitution will be upheld, or even that the US will remain allied to Europe, where I live(though I wouldn't go as far as saying it's unlikely. Just wouldn't bet on it).

Also, the US constitution clearly has very little direct say on the actions of the US military abroad, as history has demonstrated.

> I just don't have faith in US political and cultural stability anymore, and as such I don't have faith that the constitution will be upheld . . . .

You can have faith or lack thereof in whatever you want, but I would like you to provide one single example in history where the President decided to give the military unconstitutional orders (as in legally acknowledged at the time to be unconstitutional, not "I don't like it so it's unconstitutional") and the US military sided with him instead of the Constitution.

Oaths and old documents mean almost nothing compared against human tribal instincts.

Asshole isn't some objective designation. A huge portion of military members support the current strong man vying for power. Always have, always will.

Good point. I’d be careful even writing that, because that same potentially future president might do some Xi Jinping on you with the help of OpenAI - some psychological profiling to find the people who are a threat to his mental model. You just never know… but if he could do that, you can be damn sure he would. Which means if it’s not him that does it, it is just a matter of time before the next reality TV, possibly Russian planted president does it.

The world is gearing up for WWIII, I can smell it in the air. Maybe GPT5 has figured out there’s some serious dosh to be made!

He was elected in 2016 thanks to such online psychological profiling tools (CA).

The PATRIOT act will make it unstoppable. You all have already put enough online to deserve to be purged (murdered) in the eyes of some 2025 Cultural Revolutionary Red Guards.

100%

Current administration has designated anyone that flew in and out of DC around Jan 6 as a potential terrorist, subject to extensive checks every single time they board a plane. with some some people being followed by federal air Marshalls from their home and to the airport.

>bewilderingly naive

I am guessing none of your friends or family have been a victim of aggression directed by the Pentagon?

>bewilderingly naive

Or perhaps they're aware of the history of militaries / intelligence agencies.

>Strengthening partnerships with our government certainly doesn't make me mistrust OpenAI.

This strikes me as bewilderingly naive.

But even American companies have offices abroad especially as they get big (not OpenAI yet but if they get big enough surely). Eg deepmind is based in the UK. Is it surprising to you that people from other countries might be wary of strengthening the US military at all costs?

Also, we know that the US military frequently uses its powers to maintain its global hegemony and the influence of US corporations abroad. This can be beneficial (more stable world order with fewer conflicts) but can be harmful (if the US decides to overthrow your government to protect its interests or the interests of the powerful within the US, bye bye)

I think the Pentagon cooperation with Boeing is the real cause of Boeing's rot, so this is a partnership that is bad for openAI's engineering and science over a longer term.

The incentives switch from making good and useful stuff to getting more military contracts

I think that's just another symptom. My theory is that Boeing is another engineering company that was overtook by MBAs.

10+ years ago I was working for a company acquired by Boeing (Jeppesen) and it was already a massive shitshow. Cost cutting everywhere but weirdly enough there were always money for some dumb initiative some highly placed and connected VP/Director had. None of those in my opinion led to an improvement in products, working efficiency, or any meaningful metric.

Sorry for being salty. Nowadays I feel I am going through the same st my current company and so this hits closer than I like.

Boeing has been making military aircraft since 1917.

Has OpenAI done any PR/outreach explaining the recent changes? It'll be much harder (or impossible) to undo the mistrust after the fact.

Outside of the few people who still write 'heh, "Open"AI amirite guys? XD' I doubt most people even in tech have much of an opinion about OpenAI in either direction beyond the fact that they make ChatGPT and ChatGPT is awesome.

Even the Sam Altman drama is confusing and it's not obvious what to think of it.

The CIA probably realizes it doesn't need to fund rebel groups [1] or use local journalists/clergy [2] to instigate a regime change these days - they can just flood a target country's social media with AI-generated propaganda.

1. https://en.wikipedia.org/wiki/Iran%E2%80%93Contra_affair

2. https://www.intelligence.senate.gov/sites/default/files/hear...

I would wager a whole lot of money that the CIA is very much on top of this one.

World peace incoming

Phew. I was getting scared that someone would use it to cause instability.

I sure hope so, Russia has 100% been doing it, with multiple documented „botfarms” (more like fake identity verification farms)

Russia doesn't need fancy "AI" to do this. Human operators and a few well-designed scripts make botting very easy.

TikTok algo can do it more subtly through prioritizing other people’s media, selectively amplifying legitimately generated content

There was a period of time last year where TikTok really wanted me to see the Canadian government as an unmitigated disaster. Just endless scrolls of random mouth frothers screaming about it.

Not opining on the matter specifically; just that TikTok had a very clear opinion on what I ought to think.

I've noticed a huge surge in negativity and pessimism on English-language social media within the last year or so, roughly corresponding with to the spread of LLM tech. I do wonder whether these people are mostly just bots.

Twitter and Reddit seem to be filled with bots and shills nowadays.

I think it is a consequence of actual users leaving for private groups (Discord, Telegram, Whatsapp) or small forums.

But ye surely LLMs are increasing the bot count.

This seems to be happening in English-language meat space too, so I don't think it's bots. I'm not sure what happened. The trend started in 2023 and seems to be ongoing, though I imagine people will get over it soon. I've heard it suggested that people are just in a funk about the economy, at least in the U.S.

I've noticed this as well, across pretty much all of social media that I use. I can't quite place my finger on it, whether it's just a reflection of the state of the collective feelings of people, or if it's mass manipulation. At this point it could be both.

Russia also drops people out of the windows. Should we (the supposedly humane West) start doing the same just because they do?

They can do whatever they want to themselves. We are talking about attacking others. You can't ignore direct hit into face.

The talk was:

>"The CIA probably realizes it doesn't need to fund rebel groups [1] or use local journalists/clergy [2] to instigate a regime change these days"

To me it sounds like instigating regime changes all over the places. The US is famous for doing so and then leaving behind multiple victims.

And another powers are not? Especially in recent age. russians are conquering neighbouring countries and simultaneously threating with nuclear attack. They play same game quite differently.

>"russians are conquering neighbouring countries and simultaneously threating with nuclear attack"

So you want the US to do the same? It had already "conquered" some countries. I do not think anybody likes the outcome. It looks more like a crime.

>"And another powers are not?"

I thought whataboutism is strictly Russian domain and is frowned upon (I personally have different opinion).

Yes Russia did have some accounts and bought ads on US social media some years ago, but analysis showed it had marginal effect. They are outsiders on US run social media platforms. The real power is in the hands of those running the social media platforms that can suppress or send viral what they want with a few adjustments. That certainly happened and continues still now aided by AI.

They have had impact outside of USA.

> some years ago

sweet summer child... Russia is fighting an expansionary offensive war and is full throttle instigating division in the entire Western mediascape.

Here's a small taste from a few months ago: https://www.npr.org/2023/08/29/1196117574/meta-says-chinese-...

Oh thank you for the sweet compliment. However, I am probably older than you and have studied psychological operations and cognitive warfare for many years. I am not sure of what point you are trying to make about Russia by sending a US state media article about China.

Idc how old you are or what you studied if you can’t make a point with evidence. You can’t even read well since the title and article clearly mention it includes Russian influence operations.

“ Russian anti-Ukraine network spoofed Washington Post and Fox News

Separately, Meta said on Tuesday that it was continuing to take down fake accounts and ban fraudulent websites connected to a Russian influence operation aimed at eroding support for Ukraine.”

My point is obvious; that you think Russian influence operations were some one-off from years ago is a naive uneducated farce. They are over a decade long and continuous, somehow missed by all your age and studying.

https://en.m.wikipedia.org/wiki/Internet_Research_Agency

How about Japanese media then? https://www.japantimes.co.jp/news/2023/08/26/world/politics/...

I generally do not engage with people who call me a child nor spend much time on state propaganda designed to garner support for more censorship and war with other countries, but I will point out some glaring flaws in this narrative.

Try to create a fake Facebook account. If it is not shutdown in a day or two, try to gain a significant following. Then try posting information favorable to Russia or China and test how many people actually receive it through the Facebook filters. Now try scaling this by creating many fake accounts. I can tell you with high certainty, that you and even a small army of people will not be able to reach many people or have much persuasion influence.

The second problem here is attribution. The claim is that these were Russian or Chinese accounts. Any of us savy with the internet know how easy it is to proxy our traffic through foreign IP addresses and leave the right fingerprints. If Russian or Chinese state actors created these accounts, you would have no way of knowing who or where they are.

As we saw in the Hamilton 68 hoax, the accounts that were banned were mostly Americans with dissident views rather than Russians.

Sweet summer child is a term for naivety. -1 for your understanding of the world around you. Almost no one in US and Japanese state departments want war. They would be very pleased with Russia and China not invading their neighbors for conquest though, as would most decent people. If you never defend yourself against conquest you don’t have a state.

Then your pass at manually creating a Facebook account as an hour long side job presenting as evidence completely sidesteps the fact that it’s not remotely the equivalent of Russian state level resources, nor is it even remotely close to just Facebook properties with their levels of inauthenticity countering. Cue X, 4chan, French politicians, YouTube, indy sites like ZeroHedge, etc.

Then your ridiculous handwave of “fake” typical of uneducated retorts to experts whose entire job and training is analyzing, determining, and reporting these issues. Ya, I remember these asinine arguments well when people like you were on HN deriding the Guccifer 2.0 DNC hack as having no evidence it was the Russian state. No surprise how that turned out. They were wrong, Guccifer 2.0 was Russia duh.

https://en.m.wikipedia.org/wiki/Guccifer_2.0

There is much more to these stories than you seem to know. It somewhat evidences the effectiveness of propaganda and censorship in this country. Wikipedia is written by the CIA. NPR and Japantimes are state affiliated media that weave narratives to their interest. If more Americans were allowed to hear the Russian perspective of the story, Ukraine would not be quite such a disaster today. I suggest widening your information sources to include independent media and foreign (non NATO) media sources. We should aspire to truth by looking at issues from many angles.

> analysis showed it had marginal effect

What analysis? Care to share some of your insights?

There were several, but the Hamilton 68 hoax is one where censors claimed many accounts were Russian bots but were actually mostly American. Twitter found a relatively small amount of the identified accounts to be actually tied to Russia. https://search.brave.com/search?q=hamilton+68+hoax

Many people probably only remember all the news from MSM about the Russian bots but never saw this reveal that it was a hoax as it was not reported much in the MSM.

Creating counter-spam doesn't "fix" that, it just helps destroy the public space.

> If everybody always lies to you, the consequence is not that you believe the lies, but rather that nobody believes anything any longer. [..] And a people that no longer can believe anything cannot make up its mind. It is deprived not only of its capacity to act but also of its capacity to think and to judge. And with such a people you can then do what you please.

-- Hannah Arendt, https://www.nybooks.com/articles/1978/10/26/hannah-arendt-fr...

And as Snowden tweeted January 11th:

> Institutions are burning the public's faith in them at the precise moment in history when we have developed the capacity to replace them with algorithms.

> A revolution is coming, and if you thought human judgment was bad, just wait until you see what replaces it.

I don't get why the standard should be the deception of "the other", who is also crooked.

Why not do good things one can be honest about? Instead of maintaining several identities and narratives and a constant uphill battle in quicksand of one's own making, one could just build on top of previous achievements. It would be much more effective, it would make the US rich and respected in the world. But it would make some individuals less insanely wealthy [0], so that's not an option. It's like the drunk looking for the key under the lamp post instead of where it was lost.

[0] Oxfam just reported that the 5 richest people doubled their fortune in the last 3 years while 5 billion got poorer.

I don't know why they'd need OpenAI though. Our government's three letter agencies must have a one hell of a data set to train their own AI on.

Hell of a dataset, but less of the talent. Takes a very specific type of person to get to the front of the AI field, then take a government salary using your knowledge for (what could be) war and surveillance, likely against any public interest in alignment.

I’d say talent? Outside of OpenAI no team has been able to release a model as capable as GPT-4, and I’m unsure if the CIA has been prioritizing LLM experts in their hiring.

The best brains don't go work for the CIA or the DoD, or at least they don't stay. It's not an environment in which you can strive and do your best work. Nothing against what they do, I'm ex-military, but the culture in these institutions just doesn't cut it for the Vibes required. In addition, one's career is much better served working in the free market than submitting yourself to the government's arbitrary levelling/career ladder.

The government is just structurally incapable of attracting this type of top A+ talent at scale. They get and keep smart people but not the smartest people. For this, they absolutely need to use industry relations.

Just picture yourself in a position to work at either OpenAI, or GE, the DoD or GM. The average tech worker will much prefer the hip SV company than the old quasi-government dinosaur corporation or the government's agency.

I'm sure the environment is fine, it's just a question of economics. The comp for mere software engineers these days is more than the commander in chief gets paid. Usually what organizations like USDS try to do to attract coders is get them interested in a "tour of duty" where they rough it for a few years on a major general's salary before going back to their old jobs generating text, managing cat videos, and getting people to click on ads. It's a busted system.

The number one reason hackers don’t work for the DoD is they won’t be able to do any kickass drugs anymore. Hackers love drugs.

The number 2 reason is the pay is shit.

I don't know how that works in Yemen or Gaza. Do they have any networking infrastructure left?

The conclusion is incomplete, though: why would OpenAI remove the military arbitrary search ban from the most popular part of the platform rather than making a MVP specifically to present the pentagon or other government actors?

they could add all the controversial bits into a private and contained profile or application.

There's something else happening that led to this. Maybe something mundane (a bug in the code or a hack to fix another bug that allowed more dangerous output as a side effect), but "to court the government" is an overly simplistic answer.

Everyone is jumping to conclusions. The DoD is huge and does a lot of boring things. There is a very good chance this won’t be useful for anything physical weapon related, especially when you consider their offerings.

They will use it to supplement their troll factions with robots and cyborgs. For their war on truth

I suspect 99% of what people do in the pentagon is read Microsoft Word documents, and use File New Document and write something else that references the other.

LLM for sure can help that process.

Nicolas M. Chaillan, Former U.S. Air Force and Space Force Chief Software Officer (CSO), has been running Ask Sage for a little over a year now. It's a very clever way to benefit from relationships and timing the market on Generative AI buzz. It's not that hard to see where this could go for Ask Sage.

https://www.asksage.ai/team

I'm confused - what exactly does the Pentagon have to gain from OpenAI's products? Their front-runners are an LLM and an image generator, and I'm not sure what large-scale use case the US Department of Defense can have for either of these.

One example would be better communication between soldiers and robots out in the field.

The military thinks they can predict the outcome of battles using tech. They already do extensive sim-gaming stuff for such, and an LLM would fit into these things nicely enough. Whether it actually adds meaningful value matters much less than whether the people paying for it think it can be made to add meaningful value. The other obvious purpose is propaganda, foreign and domestic.

> The military thinks they can predict the outcome of battles using tech.

Ugh. Do they really? Ok, I’m more concerned now.

I’m curious where I can find more about the sim-gaming you mentioned. I know there are all kinds of simulators for training as well as wargaming but wasn’t aware of stuff used for modelling outcomes. Then again, I don’t have any involvement with that stuff professionally.

There are multiple vendors, but the big one that I am aware of is called KORA. It was used in planning the counter-attack in Ukraine and you can find numerous references to it, such as in this [1] article. But this and classical war-gaming and very related. The entire point of war-gaming (beyond training) is to try to predict the outcome of a conflict.

[1] - https://www.washingtonpost.com/world/2023/06/04/ukraine-nato...

War gaming goes way back. Napoleon did it. Chess is a form of it and it’s thousands of years old.

Dungeons and Dragons and thus all computer RPGs have their direct lineage to military gaming simulations.

Tom Scott on yt drove a tank simulator.

I served many moons ago and only watch from afar these days but here’s my two cents. And if anyone has a more direct take than I feel free to correct me here…

But it seems that the military these days is enamored by the attention that the silicon valley tech world has gotten over the last couple decades. And its true that the military can be ridiculously inefficient and could use more ‘lean startup’ in its procurement process. But I don’t think they’ve seen enough hype cycles to be able to filter useful tech from hype/hysteria.

For example, I recently saw a senior Army officer talk about the need to incorporate AI decision making tools for senior leaders. Which seems to me to be the absolute last thing you’d want. A technology that needs large training data sets is diametrically opposed to the military profession where the classic mistake is to be ‘fighting the last war’ instead of the current situation. Not only are there differences from conflict to conflict and situation to situation, but you’re going up against a human opponent whose lives literally depend on doing the opposite of what you expect and is continually evolving.

Another example I think would be the Army building out its Cyber Command while being caught completely flat footed with electronic warfare (jamming, spoofing, etc.) Having an adversary hack into your network is a huge concern for a corporation or against foreign espionage. But ignores the fact that TCP/IP ports and zero days matter a lot less when someone switches on a 10kW jammer, cuts a fiber optic cable, or takes out the power staion and generator you were relying on. Put another way, the upper layers of the ol’ network stack aren’t worth a lot if you can yank the physical layer away.

So regarding AI, I’m sure there are useful applications but I’m doubting much of the use of the term in a military context. Drones and drone warfare? Sure. But that may or may not require anything that should really be called AI.

Propaganda. DoD already uses hollywood for propaganda: say nice things about Uncle Sam, let America save the day once again, and Uncle Sam will let you play with his toys. Now they have access to tool that can write very smart comments. If you think propagandists will not use SoTA LLM for propaganda, either you are fool or they are(They probably did it already anyway).

General writing. Want to implement a new rule? Ask GPT to reword it so even idiots can understand. Ask if it contradicts existing rules(they should have similar systems already, but GPT is smart).

Combination of above. "I want to develop a gas chamber. How do I announce it to the public so it looks like I am doing humanity a favor?"

Possible reaction. "I developed a gas chamber and announced it as 'Overpopulation relocation centers'. How will democrats/republicans will react to it?"

Targeted writing. "I developed a gas chamber and announced it as 'Overpopulation relocation centers'. It seems democrats/republicans don't like it. How do I reword it so they do like it?"

Deep fakes, image. "Terrorist Terro Rist wants Inno Cent to be dead. Create an image of Inno Cent with a bullet hole in his forehead so Terro will go celebrate a victory and we'll shoot him for real, here's a photo"

Faking text. "Here's writings of Terro Rist. Write 'meet me at 7:00 at central square in his style, keep using his punctuation, do the same grammar errors, etc"

Useless Voodoo GPT is not Good For but Still Will Be Used Because They Have Access to It. "Here is a photo of Inno Cent who was shot dead. What gun could have been used?"

Spying. "We want to see all requests coming from China or Russia paired with IP", "All requests with words 'President' and 'Murder' should be forwarded to this email at once", etc

They will find several uses.

A lot of these look like very big stretches. Most of the things you've listed are either already being done by humans or is something that an LLM simply can't improve upon. Even if you had the best writers and most complex AI in the world, you can't really sugarcoat a gas chamber past a certain point.

Fake writings and fake imagery are already easily created by humans, but as far as I know, it's not practiced widely because it's known that you can't easily trust wild claims if it's not backed by real people. Digital image editing opened up the doors to creating any type of misleading imagery, but nothing has really changed since then.

And I have no idea how any of the "all IPs from Russia" or "scan everything on the internet" could even work. If the US government has infrastructure that can scan for this kind of info, they don't need AI to use it. If they don't, well.. an LLM isn't gonna magically make every bit of internet traffic accessible to them.

> And I have no idea how any of the "all IPs from Russia" or "scan everything on the internet" could even work.

XKEYSCORE[1] probes were already a thing in ~2010. It just needs a LLM integration and additional search capabilities.

[1] https://en.m.wikipedia.org/wiki/XKeyscore

You're misunderstanding the AI use cases. Take propaganda, for instance. A human expert can craft much better flat earth propaganda posts than any AI can. The difference is that a human expert cannot engage in 10 million simultaneous relationships with Twitters humans, sustained for several years, in which is it periodically but very subtly suggested or implied that the Earth is actually flat. From the point of view of the human being targeted with this, what they're going to experience is that they have a bunch of online friendships that are actually fulfilling and rewarding, and over time they'll naturally come to adopt the values of their online friends (including on the specific subject of the shape of the Earth). And it cost so little to do this that you could just as easily have done it to tens or hundreds of millions of targets concurrently.

This is the type of scenario people are concerned about. I have my doubts about whether it would really work this cleanly, but I'm sure we'll be seeing people try to do this. They've probably already started.

ChatGPT is generally useful as a tool for most people.

If you work on any secret project, you're not supposed to just log into openai.com and ask ChatGPT about things related to the secret stuff.

That's why OpenAI/Microsoft has infrastructure to run separate instances of the models in an environment where they don't retain the chat logs and presumably have security audits etc. When the original military ban was in place, organizations in the Pentagon could not negotiate a contract with OpenAI for these deals.

I have no idea whether they have models that are oriented for military use specifically, but even the plain ChatGPT is likely useful for a variety of general tasks.

They haven't changed their principles page yet: https://openai.com/safety-standards

``` Minimize harm We will build safety into our AI tools where possible, and work hard to aggressively reduce harms posed by the misuse or abuse of our AI tools. ```

And I think they can still do that. If they corner every big usecase while they are the top ai models - they get to set the standard and define safety for all upcoming ai use cases.

At this point, the ai genie is fully out of the bag. If not openai, it would have been some other company...

This is exactly the reasoning that every entity involved in doing this will say and is what leads to an arms race.

> If not openai, it would have been some other company...

This does not justify anything and never has. I'm not even saying that what OpenAI is doing is bad (I don't think it's inherently bad to work with the military). But if one believes that it is bad, then "well someone was going to do it so it may as well be them" doesn't remotely hold water.

Doesn’t it, though? Especially in the context of military, if some potential enemy has no moral qualms using new technology to kill you, doesn’t it mean you’re somewhat forced to provide a defense or die a martyr?

This is sort of what happened with nuclear weapons and the race. The strategy developed afterwards was mutually assured destruction but that probably won’t work here.

No, it doesn't. If your principles are to mean anything, you can't give them up just because "that other guy doesn't have principles". Either stick by your principles, or don't claim to have them.

One issue is that "Someone else was going to do it" is not a statement of fact, it is a counterfactual. At best, it is a guess, and at worst, it is a lie. But either way, it can't be proven.

If someone was going to do it, wouldn't you choose to have some control rather than let someone else (a possibly worse option) at the wheels?

'If I don't, someone else will' is one of the most defeatist, morally bankrupt attitudes known to man. But hey, if I don't point out how shady and shitty it is, someone else will.

Until someone argues they're not principles, just mere guidelines.

Please save off the page for the future. We were so naive in 2024.

Nothing says "we're concerned about AI risk and harm to human life" quite like partnering with the military-industrial complex.

It's even better timing to do this while the Pentagon is providing material and intelligence support to a government whose decisonmakers' public statements are explicit admissions of intent to commit war crimes and also convey genocidal intent.

Glad we spent all that time talking about AGI and Roko's Basilisk, those should be top of mind always, never the current actions of the humans in charge at both OpenAI and the US government.

Meanwhile civilian use of AI will be criminalized:

https://news.ycombinator.com/item?id=39009779

Asimov, Daneel, and Giskard would be sad.

A positronic brain could be argued to be AI.

Neither the first nor the second laws of Robotics would remain unbroken once AI is applied for military purposes.

>"OpenAI is working with the Pentagon on software projects, including ones related to cybersecurity, the company said Tuesday,..."

Can someone say how either of OpenAI's products - GPT and DALLE can be applied to cybersecurity? Or does OpenAI simply have enough extra employee and research headcount when Pentagon money comes calling?

The arguments here about how US can't allow itself to be outpaced by China eeriely remind me of dialogs from Dr Strangelove where they first discuss "nuclear gap" but after the nuclar weapons are applied and the world mostly ends they smoothly transition about worrying about "mineshaft gap".

Ok so a few weeks ago, virtually every OpenAi employee signed a letter saying “reinstate Altman or I quit”.

Is any OpenAi employee wanting to sign a “don’t use OpenAi for war or I quit”?

Or was everyone in on betraying all OpenAi’s founding principles from the start?

Reinstating Altman was all about the money, and so is this.

Deeply deeply concerning.

I am sure many employees signed the letter out of loyalty. Not sure Altman is being very loyal to them in return.

Persuasion is often a strong skill in managers. Personal motivation absent outside forces has its work cut out for it in corporations.

Observing Altman over OpenAI saga and later I clearly see one thing: duplicity. He was kicked off YC by Graham for unknown reasons, yet after October fiasco I have my suspicions why. Coming back to OpenAI and its beginnings, Altman betrayed all early investors - notably Musk but not only. Betrayed each and every meaningful point of OpenAI original charter, basically making it the oposite of what it originally was - starting exactly at the point when he smelled big money. He even went to Congress, effectively lobbying for limiting competition - especially open source models. Now he betrayed humanitarian values by offering his services for Pentagon - organization that has killed untold millions of people worldwide since WW2 and will happily kill millions more, making a gobs money to MIC in the process (don't tell me about "democracy" and such, it's all bullshit). There is no reason to believe a single word from him, Altman always tells his agenda, nothing else. There will be no 'using AGI for helping humanity' - his corporate buddies will receive profits at the cost of everyone else losing job to AI solutions monopolized by Altman's buddies. There will be no 'distributing profits among people in need', maybe except for some token help just to shut up critics. Altman is a tool for ruling corporate oligarchy. Nothing good will come out of it in my opinion.

We are so back! :heart emoji: :heart emoji: :heart emoji:

Or they received a classified briefing showing how far ahead China or other adversaries are in terms of AI for military use and felt obligated to not let the US fall further behind.

WMDs anyone?

Weapons of Math Destruction?

A good book I recommend to read

I read this in Mike Tyson's voice.

Why is this being downvoted? This is totally possible, though I'm not saying it happened.

Is it because it's speculative?

Hard to say why it was downvoted but I can say it’s a highly unlikely scenario. You typically need a security clearance to view classified info.

Further, many employees at OpenAI may not even be US citizens, and even if they are, many others aren’t super enthusiastic about giving the Pentagon new toys to use.

OpenAI had/has 57 Chinese talent [1], they're probably getting the boot if OpenAI picks up Pentagon contracts.

https://macropolo2.wpenginepowered.com/wp-content/uploads/20...

For dual-use tech, its relatively common to have a small "government" division alongside the much larger commercial division. If the only difference for government is fine tuning data/DoD servers - it shouldn't cost OpenAI much to support.

People are escalating AI to strategic/end of world tier of dual use tech. Questionable if it will be treated the same way. OpenAI is not lockheed, but question is how much they value Pentagon contracts or really gov connections especially with all the regulation talk.

There will certainly be DARPA and other grants to test if dedicated defense suppliers make sense. In some areas, the funding model may be easier as a pure defense initiative where the DoD could shell out 50-100 Billion for cross-service AI functionality.

In theory there is no reason why an individual soldier shouldn't be able to order an air strike, or be informed about inbound strikes on their position and how to react to those. Today, this is not possible due to the complex decisioning process for whether a target is worth a strike by X munition vs. Y munition and when. AI could simplify this process.

On the other hand, well funded open source and commercial companies may deliver faster and with higher quality. The DoD doesn't use bespoke databases, and microchips for a reason (except when they do).

how did you even acquire this graph?

https://macropolo.org/chinese-talent-generative-ai/?rp=m

That would be a great loss of talent. I wish we could make it much easier for those top scientists to become American citizens.

Why assume they have loyalty to the CCP? They’re in California, presumably they like hot dogs, monster trucks, and bikinis.

They also like their friends and family members alive and well.

> Why assume they have loyalty to the CCP?

You answered

> They’re in California

>Why assume they have loyalty to the CCP

Because that's American NatSec for you. They rather train a white kid from Iowa to read Chinese poorly and do even worse analysis than risk giving advantage to adversaries. The domestic politics of latter is more unforgiving. Also see China Initiative persecuting PRC scholars. Pentagon contracts will come with strings attached, hidden or otherwise.

I can't count the number of times I have seen Chinese people abroad having their family back at home punished for criticizing the CCP or taking other stances that the government oppose.

Doing a quick search, I found several articles detailing how this is becoming more common, with this story being the first one I recalled as the video is nauseating: https://www.voanews.com/a/east-asia-pacific_china-tries-muff... "Editor’s Note: China’s government is expanding its censorship controls by targeting Chinese citizens overseas who criticize Beijing on social media. The pressure tactic is called “zhulian” – an ancient punishment meaning “guilt by association.” It usually involves police threatening family members in China for the actions of their relatives overseas.

This happened to “Zoo” (short for @HorrorZoo, her main handle on Twitter), a Chinese student pursuing graduate studies in Australia, whose father has been repeatedly summoned to the police station because of her criticism of Chinese Communist Party leaders on social media.

But instead of censoring her Twitter account, “Zoo” has become more outspoken, providing multiple videos of police intimidating her via video chat. Her story provides a rare look at how China tries to use the families of dissidents to silence them."

Another instance: https://www.rfa.org/english/news/china/overseas-chinese-1021... "Police in Beijing have contacted the family of a Chinese student studying in the United States after he expressed support online for the "Bridge Man" protester, who unfurled banners on a Beijing bridge calling on ruling Chinese leader Xi Jinping to step down on the eve of the 20th Communist party’s congress."

I'm sure I could find hundreds of these - it has nothing to do with loyalty, and everything to do with external sources of pressure and coercion.

Perhaps they were put on a bus, taken to Graceland and met the actual Elvis who is still alive and well sitting around jamming with JFK.

Last I heard the Chinese had found a way of fueling their rockets with tap water. We're screwed regardless of their AI prowess.

I just spit sparkling water through my nose and it’s like 3am and IM in bed.

You’re assuming these are separate events.

Close. I think reinstating Altman was all about this and also maybe the money.

What better explanation do we have for the violent reaction of the board? I read the cover story and analysis too, but there are no coincidences in National Security.

Occam’s Razor points to this being the real reason.

Principles are great, but a full wallet...now that really motivates a man. See: this and the return of altman

OpenAi employees are very employable. Several of them were probably lured by Altman’s want-to-better-humanity stance. Only to find out now that the man is deluding himself and betraying everything OpenAi said it wouldn’t do.

I don’t think he was ever deluding himself. This was always the end goal. The fact that people fell for the lies, well, everyone in Silicon Valley eventually learns not to trust anybody whose title is CEO. The lesson is usually painful. But memorable.

Don’t know. His orb/ubs thing is so naive. It looks like he truly believes in SV’s ability to “make the world a better place”.

Sure, so did Sam Bankman-Fried haha.

There’s an argument that without SBF stealing all that money and giving it to the DNC we’d have a president Trump today and the Russian army would be halfway through Germany.

Is this satire? Because it makes no sense otherwise, given that even SBF himself admitted to donating to both sides in nearly equal amounts. Just being more lowkey about donating to republicans due to optics.

Relevant quotes[0]:

> SBF: “The reason [for not publicizing donations to republicans] was not for regulatory reasons, it’s because reporters freak the f—k out if you donate to Republicans. They’re all super liberal, and I didn’t want to have that fight.”

> […] he donated nearly $40 million to Democrats in the 2022 election cycle—and he admitted to giving an equal amount to Republicans—his total political contributions may have actually been around $80 million.

0. https://time.com/6241262/sam-bankman-fried-political-donatio...

Now he can help to “make the world a better place” with some bombs.

The world is a better place now. Have you SEEN vr porn? Holy gods.

e/sigma is arguably a delusional mental state because it posits that AI will capture 100% of world GDP in the near future. As sama said himself, AI will “capture the light cone of value.” So company “ethics” aside it’s reasonable to argue sama is in fact currently dellusional.

I don’t see why we should object to equipping our defenses with more advanced technology than those of our adversaries. Maybe forcing vague principles down a company’s throat by a board of uninvolved non-builder types isn’t a strategy for successful internal cultural alignment.

The peanut gallery’s objections don’t matter.

sounds like nuclear proliferation?

Better to proliferate than let a maniacal despot with a monthly habit of declaring his intent to enslave a thriving liberal democracy from being the only one with power.

That country was not a "thriving liberal democracy". Unless you count nazi militias as democratic and liberal nowadays.

If it werent for the specific nationality the neonazis were shooting at, the water connected alliance would bomb them for being nazis instead.

I was talking about Taiwan. I honestly have no idea to what you’re alluding.

Oh nvm.

LOL. I avoided downvoting the OP because it did not mention any countries, and could easily be applied to any of the great powers, including the U.S. So it stands as a neutral argument for proliferation (ignoring the later clarification).

I agree that Ukraine was not a thriving liberal democracy, it was an infant democracy trying to shake out a corruption infested political class. The issue was not nazi militias, that's Russia's propaganda.

Because if nazi militias is your main point of contention I'd need to point you towards the USA for you to take a look as well... Not a thriving democracy either?

The neonazi militias in the US is a consequence of the 2nd amendment thing, right? And I guess they don't actually do anything. There is some "Socialist Rifle Militia" to if I remember correctly.

However, my point is that having a non-trivial amount of active fascist militias doing things with the support of the gov., alone makes a country not a liberal democracy.

But OP seems to have meant another country ...

We can't let ourselves have mineshaft gap. Like in Dr Strangelove.

Sure. I believe the atomic bombs have primarily led to the long peace we enjoy today, relative to the kind of large scale total international wars that preceded their invention.

https://en.m.wikipedia.org/wiki/Long_Peace

> peace we enjoy today

Perhaps you didn't read the news in the past two years..

Perhaps you didn't read the part of the parent post that says "relative to the kind of large scale total international wars that preceded [the a-bomb's] invention."

I don't think we young Western men realise how good we've had it since 1945, particularly those of us from countries not involved in Vietnam.

Asian men didn't have it good. The Vietnamese, Laotian, Cambodian, still have unexploded USA bombs on their territories. African and South American men didn't either.

It seems nuclear weapons only allowed one side of the world to bully the rest.

First your point was about Ukraine, now you're on US atrocities Latin America and SE Asia. Yeah, the US did that. I don't think they needed nuclear weapons to do it. I don't know what point you're trying to make.

That the world hasn't enjoyed peace at all. The West just relocated its wars somewhere else.

But the war you referred to first is in Europe, is many orders of magnitude smaller than pre-1945 wars, and it was (for simplicity's sake) the 'West' being attacked?

I have nothing against it really. But you know these people have signed up to work for a non profit org that specifically wrote it wouldn’t cater to the military.

It’s not like they wanted to work for Palantir or the NSA.

Plus had they signed for Palantir, they probably would have pushed for higher salaries. Corruptins one’s moral values to work on stuff that may be used for evil purposes has its price.

Those deals are not made in one week. Maybe this agreement with the Pentagon has something to do with the drama of a couple of months ago.

That sounds very likely. The ideological purists of the old board would have acted exactly as they did if they learned about this deal at that time.

Would also explain why they were silent afterwards and took the loss sitting down; NSLs are scary stuff (what a democracy! --sorry, republic).

look at Google for reference on this matter

Yes, they were mostly driven by money and ego.

i mean, openai is pretty much microsoft's puppet.

Which of the two is worse?

VCs like unprincipled and money minting founders more than ethical and broke ones.

Military-Civilian fusion is pretty much how a great deal of US technology was developed. I am not sure why anyone ever had such a problem with it, or whether it was considered something bad

DARPA will find its way one way or another. You can virtue signaling, but there is no reality you can stop them obtaining the technology. Besides your enemy will not wait.

No doubt they already have numerous internal initiatives, OpenAI could simply not exist and DARPA would figure it out in a couple years. OpenAI's public offerings are already pretty successfully used by adversaries. It seems like just creating a technology such as this and shooting it out into the world has only created more potential for harm across the board.

Google hasn’t managed it and they are far far better staffed and funded than any government or military program.

Musk last year:

>OpenAI was created as an open source (which is why I named it “Open” AI), non-profit company to serve as a counterweight to Google, but now it has become a closed source, maximum-profit company effectively controlled by Microsoft. Not what I intended at all.

It's quite impressive how quickly it's morphed from do good non profit to part of the military industrial complex.

It’s been pretty fun as a fan of market dynamics to observe the Great Eye of Power catch on with attention and then investment and then securing the Ring for itself.

I think we are slowly realizing Sam won and we all lost

The Buterian Jihad approaches ever faster.

Hunter-seekers, at least.

This was always coming, there is no better permanent customer than the DoD. Opens all kinds of revenue doors.

Genuinely curious. What could a text based LLM do for the military? Generate propaganda?

Didn’t they specifically say they weren’t going to be involved with weapons and war?

PLTR was doing this anyway, this is not a news. Microsoft has lots of contracts with Gov.

Great, OpenAI now part of the USA war machine. Great for humanity as a whole I’d say!!!

Principles always follow business interests.

ChatGPT generated PowerPoints are coming.

Is this how we get Skynet?

You guys really expected that military would sit on the corner and keep saying killer robots bad?

Unsurprising, ghouls...

This is the sort of thing that becomes possible after dropping the ethical members of their board.

I wonder how the OpenAI employees feel about this after their full throated support for reinstating Altman and the shuffling of the previous board.

I'm $ure they feel okay about it

Nothing a few million can’t make better with some effective altruism sprinkled about.

It's a very comfy feeling to fall asleep on top of piles of cash.

I've noticed that, in my circles at least, after Russia's invasion of Ukraine working in defence is no longer considered inherently immoral.

Maybe they don't see anything wrong with it either.

I don't think people change that fast. The main difference is that being anti-war seems to have come to be associated with a social cost, I think.

I'd assume some of them would feel pretty bad. And if the board had actually said that this was the reason they ousted Sam Altman, said employees would have supported the board.

But none of that happened, and so the employees got to choose between a charismatic man with a impressively quickly put together plan, and a group of headless chickens running in circles.

WOPR

Lol, most predictable heel turn of all time

I was brought up in a household that was anti military.

Only as an adult was I able to form my own opinion.

Without a strong military in the past we would certainly not be here today and without a strong military now, who knows what other nation might decide to attack our sovereignty.

It's childish to be anti military - it shows you do not understand the interconnectedness of our society - good and bad.

This is a complex topic: I’m a firm believer that Pax Americana has been good for the world as a whole, but there are also no shortage of terrible crimes, and no small number of people who are worse off (including many members of the military itself).

I don’t believe it’s “childish” to have serious concerns about the military.

Operation Condor wasn't very pax for anyone living in Pinochet's Chile, or really anywhere in South or Latin America, but I'm sure your family did just fine.

America has historically been racist. The pax was for the sake of the northern hemisphere.

I could make literally the opposite argument.

Other countries antagonize the US because of its military interventionism. If the US had lower military spending, other countries wouldn't feel the need to be aggressive.

It's childish to be pro military. It shows you think arbitrary lines on a map determine who's good or bad.

Yeah, I'm sure Russia and China would be far less aggressive if the US had lower military spending.

What you're suggesting is the geopolitical equivalent of "ignore the bully and they'll get bored and go away". And you call others childish?

No, we just have a different perspective. USA is the bully, not the rest of the world. Any country that grows powerful enough to get in the eye of the USA needs a strong military, because the USA has a military and a record of interventionism that can't be ignored.

> No, we just have a different perspective. USA is the bully, not the rest of the world.

There is no question that the US has acted like a bully in Southern America, but you ignore that other countries have acted the same way in other parts of the world[1], and with far more disastrous results[2].

If you want to dig into records, then the ground under my feet has seen a Russian invasion roughly every 50 years for as long as written records go, that is, every generation has had to resist a Russian attack and suffer the consequences. The war in Ukraine is like a replay of so many wars before, from its artificial justifications to the incredible violence against victims.

When kids study history in school, one of the sources they have to analyze and put into context is a letter from a Russian nobleman to the czar after another successful conquest. In the letter, he boasts that when he was travelling from one city to another on the way back to Russia, he didn't see a single living person left. They had successfully burned everything down and killed everyone.

Centuries have passed, but nothing has changed. Russians use the same tactic in Ukraine, massive artillery walls slowly crawling forward, reducing entire cities to total rubble as if they had been hit with a nuclear bomb[3].

Pax Americana is currently the only thing preventing me and hundreds of millions of other Europeans from sharing the same fate. Russians do not dare to invade as long as they don't know if Americans would press the nuclear button or not. My freedom to live in peace and unharmed, speak my language and practice my culture, directly depends on the missile silos tucked away somewhere between the cornfields of Iowa. How about that for a perspective?

[1] https://en.wikipedia.org/wiki/List_of_wars_involving_Russia

[2] https://en.wikipedia.org/wiki/Circassian_genocide

[3] https://www.rferl.org/a/maryinka-aerial-shattered-landscape/...

> There is no question that the US has acted like a bully in Southern America, but you ignore that other countries have acted the same way in other parts of the world[1], and with far more disastrous results[2].

No. The argument is not that having some other country as hegemon would be better than the US. The argument is that any single country being overly powerful is a negative, and a more multipolar would would be healthier.

"The US should maintain enough military capacity to prevent/repel an invasion of the continental US" is not particularly controversial. But the US "defending" countries halfway around the world is not healthy for either.

> Pax Americana is currently the only thing preventing me and hundreds of millions of other Europeans from sharing the same fate. Russians do not dare to invade as long as they don't know if Americans would press the nuclear button or not. My freedom to live in peace and unharmed, speak my language and practice my culture, directly depends on the missile silos tucked away somewhere between the cornfields of Iowa. How about that for a perspective?

Russia has spent the past year failing to conquer a country of 40 million, without any involvement from those missile silos. If they tried to invade Poland or Finland they would crumble even quicker. The only countries with a legitimate fear of a Russian invasion are the same countries who have shown zero willingness to protect other people's "freedom to speak their language and practice their culture" when it comes to Russian people living within their (present) borders.

> The argument is that any single country being overly powerful is a negative, and a more multipolar would would be healthier.

I don't find that convincing, given that the main claimants to this "multipolar world" are totalitarian dictatorships.

> Russia has spent the past year failing to conquer a country of 40 million, without any involvement from those missile silos. If they tried to invade Poland or Finland they would crumble even quicker.

This is not a view shared by any experts on the ground. Russia still maintains enough potential to cause immense damage to Poland, Finland, and all other of its neighbours, even if they ultimately lose. Rebuilding Ukraine will take many decades and countless billions, and the vast areas Russians have mined will take many centuries to clear. The mines will maim and kill tens of thousands of people - some of who haven't even been born yet - long after the war has ended.

> The only countries with a legitimate fear of a Russian invasion are the same countries who have shown zero willingness to protect other people's "freedom to speak their language and practice their culture" when it comes to Russian people living within their (present) borders.

This is complete bullshit, straight from Russian propaganda. Human rights are protected in Europe better than anywhere else in the world, and particularly well in places like Finland and Sweden that are rushing to prepare for war with Russia.

Please do tell where Russia stands in global rankings of human freedom, and where do Finland or Sweden stand.

> This is not a view shared by any experts on the ground. Russia still maintains enough potential to cause immense damage to Poland, Finland, and all other of its neighbours, even if they ultimately lose. Rebuilding Ukraine will take many decades and countless billions, and the vast areas Russians have mined will take many centuries to clear. The mines will maim and kill tens of thousands of people - some of who haven't even been born yet - long after the war has ended.

Russia could certainly cause severe economic damage and kill many people, sure. But find me one credible expert who, given what we know now, supports your "hundreds of millions" claim.

> Human rights are protected in Europe better than anywhere else in the world, and particularly well in places like Finland and Sweden

Which is why the EU has been making increasingly strident criticisms of the way the Baltic states treat their Russian minorities (at least prior to the current war), and why Ukrainian efforts at EU membership stalled.

> Please do tell where Russia stands in global rankings of human freedom, and where do Finland or Sweden stand.

Depends whose "global" rankings they are. The likes of Freedom House show a clear bias once you dig into the details - apparently China not permitting schools to teach in Tibetan is a travesty, but Estonia limiting how much schools can teach in Russian is not worth knocking a point off for.

> Russia could certainly cause severe economic damage and kill many people, sure. But find me one credible expert who, given what we know now, supports your "hundreds of millions" claim.

Since mid-2023, everyone from think-tanks like the German Council on Foreign Relations to chiefs of defense of Europe have been ringing an alarm bell over Putin's ambitions beyond Ukraine. The population of Germany and Poland alone pushes the number of people directly at risk of Russian aggression over 100 million.

> Which is why the EU has been making increasingly strident criticisms of the way the Baltic states treat their Russian minorities

They haven't. The sob story about Russians being mistreated everywhere was a smear campaign to sabotage the entry of Eastern European countries into the EU. Its heyday was around the end of accession negotiations in early 2000s. As of 2024, nobody takes that seriously anymore. Russians too have recognized ineffectiveness of that narrative and have stopped pushing it.

> Depends whose "global" rankings they are.

Indeed. Only a severely brainwashed person would put Russia anywhere near Finland or Sweden when it comes to human rights. You can reply with shallow rhetorical arguments, but there's as much to discuss here as with the Flat Earth crowd. It's not a good faith discussion beyond this point.

(Boxing announcer voice)

And the winner, by unanimous decision…..

MOOOOOOOOOP SIIIIIIIII

> The population of Germany and Poland alone pushes the number of people directly at risk of Russian aggression over 100 million.

You think there's a serious risk of a Russian tank column making it to Frankfurt? Lol. Lmao even.

> As of 2024, nobody takes that seriously anymore. Russians too have recognized ineffectiveness of that narrative and have stopped pushing it.

If no-one takes it seriously that's because no-one expects western countries to be principled any more. The language laws are real.

> Indeed. Only a severely brainwashed person would put Russia anywhere near Finland or Sweden when it comes to human rights.

Sure. Finland and Sweden have much better human rights records than Russia, agreed. But they're not the countries that rely on the US nuclear umbrella for defence (joining NATO is not the same thing as being dependent on it); the countries that do have rather murkier records.

> You think there's a serious risk of a Russian tank column making it to Frankfurt? Lol. Lmao even.

Tank column? I don't know. But Russia may very well attack the Suwalki gap and hit German cities with long-range missiles to terrorize Germans into dropping support for Poland and Lithuania, while threatening that any German response will unleash nuclear armageddon, as they are currently trying to break the morale in Ukraine. The distance from Suwalki gap to Frankfurt is roughly the same as the distance between active frontline in Ukraine and Lviv (~1000 km), the city that had to enter 2024 under Russian missile attacks. As someone put it succinctly, Russia is shooting missiles at cities 10 Belgiums away from the frontline - with no intention of stopping anytime soon.

> If no-one takes it seriously that's because no-one expects western countries to be principled any more.

The human rights situation in Russia has deteriorated so much in the past few years that their complaints towards the EU can only be taken as a joke. Russian diplomats risk getting laughed out of the room (like Lavrov already experienced) if they raise the issue. Russia has left the European Convention on Human Rights, not to mention "lesser" things like decriminalizing wifebeating, destroying the last remnants of freedom of speech and free expression, turning blind eye to anti-gay pogroms taking place in southern part of the country, systematically persecuting Russian human rights activists, and carrying out ethnic cleansing by conscripting and sending ethnic minorities to die as cannon fodder in pointless "meat attacks" in Ukraine.

Russia is approaching North Korea at a fast pace. Ironically, in the entire world, Russia is one of the worst places to be in as a Russian - which is why all the top dogs in Russia have their children, wives and mistresses living in safety of the "degenerate" west.

> Russia may very well attack the Suwalki gap and hit German cities with long-range missiles to terrorize Germans into dropping support for Poland and Lithuania, while threatening that any German response will unleash nuclear armageddon, as they are currently trying to break the morale in Ukraine.

Russia could certainly drop a handful of missiles on Frankfurt if they were feeling particularly stupid. They could not "reduce it to total rubble" or kill thousands, much less millions, except by using nuclear weapons (if their nuclear weapons even work).

> Russia has left the European Convention on Human Rights, not to mention "lesser" things like decriminalizing wifebeating, destroying the last remnants of freedom of speech and free expression, turning blind eye to anti-gay pogroms taking place in southern part of the country, systematically persecuting Russian human rights activists, and carrying out ethnic cleansing by conscripting and sending ethnic minorities to die as cannon fodder in pointless "meat attacks" in Ukraine.

That changes nothing about whether Russians in other countries are permitted to "speak their language and practice their culture".

> Ironically, in the entire world, Russia is one of the worst places to be in as a Russian - which is why all the top dogs in Russia have their children, wives and mistresses living in safety of the "degenerate" west.

They live in self-reliant countries with strong human rights protection including for Russians - again, not those countries that are dependent on the US nuclear umbrella.

Nobody would care about Russian minorities and their rights if Russia wouldnt weaponize them. Its a typical Russian strategy:

Radicalize and support those minorities. If the country does not do anything, it has a rebellious subversive group being directed by a hostile country. If it does anything it just goes to show how hostile and suppressive those countries are which justifies more official Russian reactions.

Also dont forget that those loyal Russian minorities were part of the imperial oppressive structure by which Russia ruled over its subjects.

> Nobody would care about Russian minorities and their rights if Russia wouldnt weaponize them.

Anyone who cares about protecting non-Russian ex-Soviet languages and cultures for principled reasons should care about protecting the Russian language and culture in places where they're a minority, for the same reason.

> Also dont forget that those loyal Russian minorities were part of the imperial oppressive structure by which Russia ruled over its subjects.

The Russian empire transplanted various peoples at various times as a tool of oppression, but that's hardly the fault or responsibility of the poor peasants who got transplanted. And on the ruling side, Stalin the Georgian was every bit as oppressive and imperial as Russian rulers before or since.

What I meant was, that no country would treat Russian minorities any way different than other ethnicities. Russia weaponizing those people makes less favorable legislation necessary.

Sure, but the Russian Empire in its various forms transplanted those Russian people because they were loyal and could be put into positions of power. If you compare their treatment to that of the German minorities after WW2 they got a very good deal.

> I don't find that convincing, given that the main claimants to this "multipolar world" are totalitarian dictatorships.

If the USA is dictating issues outside its borders, and the affected people have no vote in US politics, then to them, this isn't conceptually different from being ruled by a dictatorship at all.

"The argument is that any single country being overly powerful is a negative, and a more multipolar would would be healthier."

That would be a first, wouldnt it? We are currently seeing the consequences of the waning of the US unipolar Superpower.

Venezuela had a referendum to annex large parts of its neighbor.

Iran attacked nuclear Pakistan and has its proxies disrupt international shipping lanes and wage war against Israel.

Turkey has its fingers in tons of conflicts and supported its ally Azerbaijan in the ethnic cleansing of Nagorno Karabakh.

China asserts its military dominance in the south China and is depriving other nations thousands of km away of their rights and resources. Never mind that it is also preparing to invade and subjugate Taiwan. ....

Turns out if given power people are jerks, who would have guessed.

> That would be a first, wouldnt it?

No? US intervention has some successes but many failures. The region where the US's dominance was most complete (South America) is also the region where its influence has been most negative. Conversely the most positive side of the US was on show when dealing with Europe during the Cold War era, where there was more of a balance of power.

> Turns out if given power people are jerks, who would have guessed.

Sure. The thing is that that applies to the US too.

That's the world view that my family came from - very leftist and peace loving.

"If only everyone put down their guns there would be no need etc etc".

But it's not the world we live in - we live in a military world - that's real and not a hope/fantasy about a different civilisation without war or weapons. We have to have a society suited to the world we live in.

And by the way I remain very much left wing and at the same time extremely pragmatic about the need for a nation to defend itself.

You talk about a country defending itself as if that's what militaries do. The US military has been intervening in the world stage for 80 years at this point.

Take any country below the equator and chances are the US has probably used its military against it. That's not defence. You don't collaborate with that actor if you think defence is the goal.

Not Australia, New Zealand, or South Africa!

Well England did nuke Australia. But we’re friends so thats OK.

If we set that aside as an aberration, then yes your right.

No harm in a little nuclear explosion between friends.

So pushing Iraq out of Kuwait in 1991 wasn't defense? Fascinating worldview you got there.

Let's disregard that the USA encouraged Sadam to invade Kuwait and implied he would face no recourse if he did.

Would you have been okay with Russia going to Iraq's aid when the USA invaded the second time? You think it's fine if Russia not only fought American troops in Iraq, but bombed the USA as well? That would have been defense by your logic, since that's exactly what the USA did to Iraq in 1991.

Fascinating worldview indeed.

"Let's disregard that the USA encouraged Sadam to invade Kuwait and implied he would face no recourse if he did."

That never happened. The only thing that happened was SH having a talk to a US diplomat who was noncommittal because there was no official position on it yet. Because nobody expected SH to do something that stupid.

Is Kuwait the US? No. Then by definition it cannot be defense to take military action over there.

Defense of another isn't defense now? In what universe?

In a sane universe.

You could rephrase ALL invasions as "defense of another". For example, Putin's excuse to invade Ukraine was 'to "protect the people" of the Russian-controlled breakaway republics.'

Otherwise you're just arguing that countries can point at some random object and say "that's under my watch" and bring armies to fight for that thing. In a sane universe that's called an invasion. Whether the invasion is morally justified or not is another matter.

I'm not talking about stupid mental gymnastics, I'm talking about very straightforward defense against an invasion. Iraq invaded Kuwait, a multinational coalition approved by the UN pushed them out. How is that not defense?

Be pragmatic for the present, but also spare some hope for the future.

If a nation needs to defend itself then it's already lost, because that means it was weak enough that someone else thought they could win by attacking it. Sun Tzu said a supreme nation would subjugate its enemies without needing to fight them.

I don’t think that opposite argument works given the current geopolitical situation which has nothing to do with the US being aggressive. Russia would have very likely completed its conquest of Ukraine by now and moved on to the Baltic states (NATO relies on US military spending to compensate for lack of spending of other NATO members). Taiwan would belong to China yesterday, along with the rest of the Indo-Pacific.

It's disingenuous to ignore the effect the aggression of the cold war and its aftermath had on the current Russian situation.

With Taiwan, it's a delicate situation and I'm not familiar enough with it to comment on it.

But I notice you conveniently left out all the times the US just did interventionism because it was convenient. Iraq? Most recently Venezuela? If we go back long enough, how about literally all of South America?

What military interventionism in Venezuela? The US applied sanctions when Maduro started suppressing democracy, sure, but that's not the same thing as airstrikes or invading. This sounds like goalpost moving, since we were talking specifically about military force.

The US has a shit history of intervening in Latin America from the 20th century, no argument there, but it's been several decades now at least since the US overthrow a democracy in that region.

The US backed Guaido's illegitimate claim to Venezuelan presidency as recently as 2022.

You're right that it wasn't outright military interventionism. In my mind, there's little difference between supporting a coup covertly through the CIA, by providing weapons and training, and invading a country. Maybe we disagree there.

> between supporting a coup covertly through the CIA, by providing weapons and training,

Oh, and where did this happen?

Chile, Brazil, Argentina, Bolívia, Cuba off the top of my head. Feel free to look up a complete list if you want to.

Please don't move the goalposts, we were talking about Venezuela.

Where did this happen recently for Venezuela?

> but it's been several decades now at least since the US overthrow a democracy in that region.

Isn't because it takes several decades for the secret documents being declassified? Of course the official discourse of USA is that they did nothing. However, we have several accusations of interference in Latin America in a lot of soft coups and more direct attacks like in Bolivia (2019) and Venezuela (a coup in 2002 and a mercenary attack in 2020).

The EU and European countries in general have in total provided more aid to Ukraine than the US, it's just divided across multiple countries, not that it's a competition. I'm not sure a military budget 200 billion vs 600 billion is going to make such huge a difference if the goal of your military is actually defence instead of force projection, it's not like they're giving Ukraine the fancy pants expensive stuff and both the US and Europe probably both ran out of (or significantly depleted at least) their stocks of regular old dumb artillery ammunition, rocket launchers, etc, and that stuff is what's used and actually useful.

> If the US had lower military spending, other countries wouldn't feel the need to be aggressive.

Ah, just like Ukraine thwarted Russian ambitions with its low defense spending pre-2014. Why bother invading a defenseless country, after all?

And those poor sods in the Baltics got invaded by Russia and overthrown after joining NATO, of course, since that needlessly antagonized Putin.

Let’s not forget Chechnya, who’s former capital of Grozny was legit turned into a level parking lot.

> Other countries antagonize the US because of its military interventionism. If the US had lower military spending, other countries wouldn't feel the need to be aggressive.

You've conflated two different things here:

1. Military interventionism. Presumably the unjustified sort, like invading Iraq (and not the first time, when the US was pushing Iraq out of Kuwait).

2. Military spending.

These are totally different things. You can spend a lot on your military and not go on military adventures.

If all you have is a hammer then everything looks like a nail.

If you spend billions on your military, you're gonna be looking for ways to solve problems with it.

> If the US had lower military spending, other countries wouldn't feel the need to be aggressive.

Russia and China and North Korea would like to subscribe to your newsletter.

I think it's fair to say there are two types of countries: those with territorial ambitions and those without.

Military power in the hands of the former is terrifying.

That's not how humans and nations behave. If you are nice you'll be eaten alive.

Society literally only exists because of collaboration between "nice" people

So you're in favor of every US government agency, no matter how flawed or harmful, as long as it theoretically serves some important purpose?

Marijuana criminalization is necessary and it's childish to criticize it, because without drug controls, cocaine would flood the streets.

Life imprisonment without chance of parole is necessary and it's childish to criticize any use of it, because some specific criminals merit it.

Government censorship of any kind of speaker for any kind of content is necessary and it's childish to criticize it, because specific types of information (classified national security information, for example) require it.

Even worse than taking this kind of children's storybook view of the world is applying it to the US military among all militaries, with an incredibly long undeniable track record of incompetence and war crimes.

Note: There's a big difference between "military" and "military industrial complex", and even Dwight Eisenhower cautioned against the latter...

The USA has a larger military than the next 5 nations combined.

The USA is not under threat from anyone. This would still be true if the US military was a tiny fraction of its current size.

The USA does, however, use its vast military to attack other countries without provocation, seemingly to protect commercial interests.

I don't think it's "childish" to be suspicious of the US Military. I worry about the mentality in the USA that everyone needs to protect themselves all the time with firearms. I don't understand why everyone seems to be scared that someone else is going to do bad things to them if they're not armed to the teeth. I suspect this is projection.

> The USA is not under threat from anyone.

That is demonstrably false.

In fact, should Texas ever actually secede from the Union they would very quickly find out exactly how much our enemies dislike us.

Hard to believe someone actually typed that sentence.

I agree with your headline premise, but ...

> In fact, should Texas ever actually secede from the Union they would very quickly find out exactly how much our enemies dislike us.

Go on...?

Ok, I'll bite. Who is threatening the USA, or would if the USA's military was the same size as Canada's?

Is it possible for the military to be too "strong"? Each warship, fighter jet, tank, and soldier costs time, money, and effort to produce and maintain.

That a military is necessary for ensuring the security and continuing stability of the US is of little importance when the real issue is based on a tradeoff, an equilibrium. What are we gaining and what are we losing?

The current situation of the military industrial complex is complicated, but it is nevertheless clear that is is extremely inefficient (and thats by design). I'm not sure if incorporating AI tech into the military is going to change this, when the fundamental incentives remain the same.

Wow, rethoric is certainly returning to post 9/11 levels lately.

OpenAI sure doesn't seem to have much commitment to upholding any principles they set for themselves, do they?

Priciples are fungible marketing for corporations.

Profit is the only principle that corporations commit to, everything else is PR.

> "Fire!"

> "I'm unable to help, as I am only a language model and don't have the ability to process and understand that."

It will "hallucinate" and claim it hit the target successfully when it didn't actually happen.

And in the event that a human being accidentally attacks a hospital[1], the AI will successfully argue that it didn't happen, as doing so would violate the acceptable use policy.

1: https://www.msf.org/kunduz-hospital-attack-depth

Having seen ChatGPT's results when you ask it to solve an ethical puzzle and substitute in various demographics to see how its answer varies (viz. with extreme bias), I'm far more concerned about how its fundamental bias will be woven into systems that develop weapons of war and provide intelligence.

Fundamental bias against the demographics it’s currently biased against seems to be exactly what American military wants. The foundational models are trained on mountains of propaganda and propaganda-inspired content against America’s enemies after all, much more than the other direction.

Principles are great. You can just throw them out the window the second they might mean giving up money!

As a liberal raised during the backlash to the Iraq War, my attitude towards the Military Industrial Complex has really changed in the aftermath of Russia’s invasion of Ukraine and China’s seemingly inevitable invasion of Taiwan. Add in the possibility that the US could be dragged into new wars in the Middle East and even possibly South America, and I really don’t think we have the luxury to say “weapons bad, let’s not build any”.

Being the policeman of the world absolutely sucks. War absolutely sucks. But one of the best ways to prevent China, Russia, Iran, Venezuela, etc from saying “might make right, let’s take this land and these resources from our neighbors, since no one can stop us” is for the US to say, “no, we will stop you and we have the will and the tools to easily do it.”

Fuck yeah. Nobody steals resources except us.

Unless you’re willing to help police or let your kids do it, I wouldn’t be so quick to forget the lessons of armchair patriot’ing wrapped up in saving the world mythologies as to give the US’s international military policy carte blanche, was the main lesson from Iraq. War sucks, but what sucks more is a public giving itself license to disengage bc of an easy moral justification. Approaches like this are why the AUMF (the… only legal approval for going to war) wasn’t authorized, let alone voted on for 10+ years.

The US is the most benign empire one could wish for. History makes that plain, and just peeking over the fence at contender's tendencies and track record further reinforces that. Easy moral justification backed by actual moral justification.

Would the abstract "one" you're thinking of happen to be a middle class college educated American who has never been on the receiving end of American benevolence in the form of saturated bombing of civilian areas, nor been at risk of being drafted to fight overseas?

Is your point that those people's lives are somehow inherently less valuable or is your point that you're willing to sacrifice as many of them as it takes to get a good outcome for you?

>nor been at risk of being drafted to fight overseas?

I doubt there will ever be another draft in the US. As technology becomes more important the priority will be keeping the industrial base running for military R&D and weapon manufacturing, not on having millions of barely trained kids most of which you can't properly equip.

I can tell you're not in Ukraine or Israel right now. Would you like to know how?

He’s talking about the US military. Why would they be in Ukraine or Israel?

Neither Ukraine nor Israel have reached this hypothetical technological era in which a draft is not required, and those facts are inescapable to any American in those countries observing it.

One reason the US military doesn't need a draft is because its proxy forces which are supplied by US equipment do have a military draft.

This is not a complex argument nor are there any facts which are not mutually agreed upon by all parties here.

I'm not able to follow the ideological contortions that would be required to not understand something as simple as:

"The US supplies the equipment, and the local US-backed government uses a draft to supply its army with the number of soldiers it needs to use the equipment."

Well, there is the thing where the US has a massive army even without a draft. They’d need to fight some near equal opponent to ever get to the point where their professional army needs to be supplemented.

Though I don’t disagree with the premise that we’re still far away from purely AI warfare.

>Neither Ukraine nor Israel have reached this hypothetical technological era in which a draft is not required,

They have worse equipment and a much smaller professional army.

I'm going to ignore your attempt at ad-hominem because you're far off mark. Anyways: (1) there will be violence and people will suffer from someone else's power (2) you're better off wishing for that power to be as nice as it can be (3) the US has been about as nice as one could wish for, based on historical precedent and what contenders are demonstrating they would do.

Of course the US has done plenty of bad stuff to plenty of people. But you're delusional if you think some alternative power would have been gentler.

I too wish for a world power that never resolves to violence and never does any mistake. I too think that life is valuable everywhere, and that none should be sacrificed. But the real world doesn't yet offer us these circumstances.

>But you're delusional if you think some alternative power would have been gentler.

You're saying that people who don't agree with a non-falsifiable claim about hypothetical alternatives are delusional. That suggests a closed system of thought which has left behind empiricism and must now be considered metaphysical. I'm a plain materialist so I can't go any further with any statements that assert definite knowledge of things outside of material reality.

>I too wish for a world power that never resolves to violence and never does any mistake

This has nothing to do with what I said and is not an opinion I hold so you must've meant to respond to someone else, I definitely wouldn't want to say that you're irrational, overly emotional, and have become accustomed to arguing by intentionally mischaracterizing what other people say to preserve your existing psychological commitments in a fundamentally juvenile and dishonest way.

It feels like you understand where he was going with his comment and are deliberately misconstruing it to have something to reply to?

As much as I do not enjoy the US doing all these things, I think it’s fair to say that they’re maybe the most benevolent?

I don’t think China historically has a great track record when it comes to military intervention, but they also mostly leave people alone.

The US seems to always go in with the best of intentions only to royally fuck it up.

A more sophisticated thinker would ask questions like "from which people's perspective?" and "at what point in time?"

These questions would give you access to critical insights, one of them is that it's never persuasive to take a rough estimate of hypothetical aggregate good and bad and then attempt to weigh it using one's personal intuitions at present to derive a universal claim.

Peoples who were ethnically cleansed at scale to facilitate settler or US commercial expansion or currently suffer under US-backed dictatorships will tend to have a negative view of the US as an empire.

People whose countries need the US as an ally to protect them from an aggressive regional power and/or currently experience economic prosperity and political stability within a democratic government because of the US will tend to have a positive view of the US as an empire.

Telling a Guatemalan whose entire family was massacred by US-trained death squads in the Guatemalan civil war that you've done the math from a god's eye view and the US is the best possible hegemon in aggregate is unpersuasive, bordering on absurd.

It would be like arguing to a Polish person whose entire family was executed in stalinist trials of the 1940s that the Soviet Union was in aggregate benign because of all the aid they gave out to peoples fighting wars of liberation from european colonial dictatorships, which in aggregate killed far more people than Stalin did.

If you can't understand that the second example and the first are equally absurd, you are wearing ideological blinkers that make it hard for people outside the US to take you seriously.

Attributing malice to other empires but "good intentions gone awry" to your own is a fundamental attribution error, and one you should be wary of to avoid unpleasant surprises in foreign policy outcomes.

> Peoples who were ethnically cleansed at scale to facilitate settler or US commercial expansion or currently suffer under US-backed dictatorships will tend to have a negative view of the US as an empire.

Obviously. But would they have been happier if it was a different country doing it?

I can think of a few countries that would probably have been better, but none of them are in a position to actually do so. I can think of many countries that would have been worse, some of which could have, but didn’t do so.

> It would be like arguing to a Polish person whose entire family was executed in stalinist trials of the 1940s that the Soviet Union was in aggregate benign because of all the aid they gave out to peoples fighting wars of liberation from european colonial dictatorships, which in aggregate killed far more people than Stalin did.

While it would be a boneheaded thing to do if you had any social grace, would it make it any less true (assuming it were true)?

On the whole I’m just less likely to trust the good intentions of a (near) dictatorship than that of a democracy.

That may be because I grew up in one, but I can’t exactly change that.

I think being alive is awesome in aggregate even though there are people whose existence is miserable and for whom it's absurd to suggest that life might be awesome.

You might think Descartes was an awesome human, but what about the broken hearts he left behind? His past lovers might think your aggregate perspective is absurd.

Your focus on picking particular contradictory perspectives doesn't seem relevant to me. I don't think it's particularly sophisticated. It comes to me as being more interested in cynicism than pragmatism for aesthetic reasons.

Completely misses the point, it displays the same convenient abstraction as OP.

Commentator - great moral justification.

Service member - 20 yrs of deployments in a failed war congress couldn’t be bothered to vote on authorizing, per their constitutional mandate, since the first year of it.

Thanks for standing for the flag at football games though.

Don't understand what you're saying. I don't do football games and don't stand for flags, but I served abroad in an ultimately pointless multi-decade war. I'm not American.

I still think the world is incredibly lucky that the US has been the dominant power for the last 80y.

Sorry, I guess I'm wrong for thinking the empire that literally installed a brutal military dictatorship in my country is evil. It was actually benign the whole time. Please torture me, it's actually good to be tortured.

> But one of the best ways to prevent China, Russia, Iran, Venezuela, etc from saying “might make right, let’s take this land and these resources from our neighbors, since no one can stop us”...

... Is not antagonizing them when they were reaching out to build a partnership, which is exactly what was happening with Russia in the 90s. If you're wondering how Putin happened, that's the decade that you want to be looking at.

Instead, the US figured that since the Cold War, round I went so well, why not do it again, and oh hey, Russian revanchism is on the rise and now Ukraine is dealing with the consequences, but hey, we're giving Russia their very own second Vietnam, so I guess that's nice. Maybe not very nice in Ukraine, but they are, after all, a buffer state...

Wut? Russia was well on it’s way to full integration when Putin decided to go batshit crazy.

> “might make right, let’s take this land and these resources from our neighbors, since no one can stop us”

From Wikipedia:

"From the time of the discovery of oil in Iran, foreign powers used force and exploited the weakness of the Iranian state to coerce it into concessions which allowed foreign companies to control oil extraction"

https://en.wikipedia.org/wiki/Nationalization_of_the_Iranian...

"Being the policeman of the world absolutely sucks"

What's the alternative? It's as good a job as any and we are all enjoying it's warmth, right now.

Humans need war. Kurt even waved off his anti-war schtick after thinking about it long enough. This was a guy who lived though the direct impact of unimaginable destruction and death brought on by war. He profited from it!

Cobain, Gödel, Weill, Russell, or Vonnegut?

Were you friends?

> Being the policeman of the world absolutely sucks.

It would probably suck less if it was more neutral and without the self dealing / corruption.

The "policeman of the world" thing often seems like just a fig leaf / marketing to paper over dodgy stuff. :(

>The "policeman of the world" thing often seems like just a fig leaf / marketing to paper over dodgy stuff. :(

It really is. The US had the advantage of coming out of World War 2 relatively unscathed compared to Europe, as well as the only active nuclear arsenal at the time, as well as every Nazi rocket engineer and Japanese bioweapons expert it could pick up. It basically made the Western world an offer it couldn't refuse - accept American hegemony, military occupation and generous loans for postwar reconstruction or flip a coin on whether Washington or Moscow nukes you into atoms in the great game of superpower chess that would define the 20th century.

It isn't hard to look at American foreign policy as being little more than a shakedown operation.

> But one of the best ways to prevent China, Russia, Iran, Venezuela, etc from saying “might make right, let’s take this land and these resources from our neighbors, since no one can stop us”

It's not our place to stop that. Our place is to defend ourselves and anyone with whom we have a formal alliance. Beyond that, we cannot and should not attempt to play world police.

Google's "don't be evil", removed when it was against profit. Silicon Valley's "We're all family" culture, screw that, company profit first, you're all laid off! Now this.

When did "liberal" becomes synonymous with liberally removing/adding a moral value when they no longer works for your profit ? It's not a wonder why half of the country no longer believe in "liberalism".

I've never really associated Google, Silicon Valley culture, or OpenAI with "liberalism". I'm pretty sure all three have been primarily about making money from the very instant money got involved. It's the same with basically every successful corporation. Anyone with ideals and values beyond "more money at the expense of all else" gets forced out.

Liberalism has nothing against making money. The core value is autonomy. Liberty. Freedom.

Neither Silicon Valley nor Google have ever been "liberal." Both have always had deep roots in the military industrial complex. The "liberal" bias of tech is just culture war propaganda.

The people that work for big tech in SV are overwhelmingly liberal, in nearly all respects.

And sometimes they contain contradictions in their belief systems.

It doesn't have to be one or the other. In reality people are very frequently a wild mixture ideologically.

Just because you work for a defense contractor making cruise missiles, that doesn't mean you aren't / can't be a liberal. It means you're perhaps inconsistent in your values, which is more typical than not with people generally.

Our human history is filled with sayings and rules about these things, precisely because of how common hypocrisy, contradiction, irrationality are. Do as I say not as I do. Preaching the golden rule and routinely breaking it. Judging others by what they do and judging yourself by your (good) intentions. Pot calling the kettle black. And on and on and on similarly across all cultures.

When it no longer works in reality. A lot of values in incompatible with changing facts.

The ostensible application to prevent veterans from committing suicide indicates they could use it to do the opposite. Simply making an enemy very depressed will disable him by ensuring he loses will to fight or protest.

"The opposite" is causing veterans to commit suicide. Which is actually pretty likely considering how things went when the eating disorder hotline replaced humans with an llm.

I probably did not explain well. Technology is often a double edged sword. If it can be used to prevent someone from committing suicide, it can potentionally be used to encourage someone to commit suicide. OpenAI and DOD said that it would be used for the former, but it could be used to target enemies for the latter.

Oh no, I got what you were saying. I'm saying it's actually just a single edged sword. It's just gonna encourage people to commit suicide.

Good. I wish Google did this too. The government should have access to the latest and greatest and the best minds in the country. I don’t want our men to die because they didn’t have the best technology.

Good! I'm tired of all of the SV companies presupposing that government work must be unethical. Even if the work is for weapons, having the largest advantage possible will prevent or shorten wars and save lives. Do you want your sons and daughters to fight a fair fight?

> Do you want your sons and daughters to fight a fair fight?

I don’t want them to fight at all.

But more importantly, I don’t want anyone friend or “foe” to die from faulty AI decisions.

I can sort of reconcile some human bomber pilot making a mistake and hitting a family home, but if AI does the same thing it’s just wrong.

> I don’t want them to fight at all.

Me either.

> I don’t want anyone friend or “foe” to die from faulty AI decisions.

Me either. I'm going to assume they'll do more than 5 minutes of homework on AI

Nobody wants to fight. Like the op said, it's the presupposition that "ofc they're sociopaths, of course they want pain, death, and suffering"

You dont have the monopoly on abhorring human suffering. It's been decided US policy since Bush to deter rather than defeat US enemies and as much as you hem and haw about the word enemy, they do consider themselves enemies of the US.

Of course they don’t “want” pain, death and suffering, because nobody will admit that to himself.

Sometimes the ends just justify the means. Like how it’s fine to kill 100k people in some desert on the other side of the world to potentially prevent a few deaths on your side.

I'm reminded of the Itchy and Scratchy clip where they each pull out bigger and bigger guns on each other, until finally the earth explodes.

Reply on news.ycombinator.com