Obviously anyone who has used LLMs know they are not on par with humans. There also needs to be an accountability framework for when software makes the wrong decision. Who gets fired if an LLM hallucinates and kills people? Perhaps Anthropic's stance is to avoid liability if that were to happen.
We need to do it because our enemies are doing it, in any case.
Of course they have the means. Nothing technical prohibits them from blowing couple of carriers. But the price they would have to pay is way too high.
Because there are actual technical impediments why neither China nor the Russians could have blown a US carrier in the Caribbean.
I do not believe so. Not unsurmountable at least. The consequences are however far from pleasant for each side
Furthermore, you mentioned this in response to "helping Venezuela", but even damaging a carrier (something technically very, very difficult for Russia or China) would not have helped Venezuela one bit.
It'd be more technically feasible for them to help Iran than Venezuela, and even that is not particularly feasible now, other than very indirectly.
I think it would, meaning that right from that exact minute the US and Russia will be very busy and Venezuela left to it's own devices. Does not mean Venezuela would feel any better of course.
There's no effective way of Russia to militarily help Venezuela and strike any US carrier. Same with China. You haven't proposed any because there is no feasible way.
Even if they could, such action would have been followed by the US knocking Venezuela out and taking them out of the equation. A neighboring ally of an actively engaged hostile power wouldn't be "left to its own devices".
If war is safe to wage, then it just means we'll do it more and kill more people around the globe.
What genuine threat did Venezuela or Iran pose to Americans? Corporate interests don't count.
As for delivery, Iran does have missiles capable of launching a nuclear weapon at American assets in the Middle East, or American allies. Or even to just float it over on a ship.
There's reports Iran agreed to limit themselves to only medical grade centrifuges as recently as last week.
And no, Iran does not have weapons capability to reach the US, period.
They fundamentally did not pose an imminent threat to the United States. A threat to American strategic goals is not an imminent threat to the American people.
Is your claim that the deal was not preventing Iran from developing a nuke? Then why does the existence of the agreement matter either way?
Are you saying Iran would magically produce a nuke the very day the deal expired? Then why don't they have one today?
How does ending the agreement make it harder for Iran to get a nuke? How does "tearing it up" prevent anything that the agreement itself wasn't preventing?
You can't condemn one and condone the other on that basis.
Iran has both reason and were developing capability to destroy a significant part of American national security. America absolutely must prevent that at any cost.
You could argue about how the rhetoric between the states got so bad that they each threatened each other's destruction. But the fact is that they got there.
That is very boring, caveman logic.
There exists a culture for which it is an honour to kill Jews. Pretending that this culture does not exist is racism. Disregarding the differences in values of other cultures is the most disgusting form of racism - pretending that one's own culture is dominant or universal.
You seem quite concerned with the plight of the Palestinians so I'll use that as an example: Jewish people experienced the worst, most widescale crime against humans ever committed, and then a few decades later, a subset of Jewish people turned around and began doing the same (at a smaller, less industrialized scale). This demonstrates the perfect universality and programmability of a human, which includes "human culture."
Any human culture can be molded to justify any existent human action. To pretend otherwise is to engage in ethnocentrism - what you accused me off, the presumption that there's something special about your culture that prevents atrocities happening under it.
The second that makes your argument racist rather than logical (if you refuse to budge on the word "racist," swap in "prejudice" - the fallacies are the same either way) is the homogeneous angle you're applying. This should be an obviously fallacious statement: "Christianity is a violent culture that supports violence against Jewish people, discrimination against gay people, and school shootings." Why is it fallacious, though? I know lots of christians that have done all of the above, proudly tying it back to their religion. You see my point, right? You would, presumably, never walk into a room of white people and assume they all share identical values - do it in America, half probably are tearing their hair out in frustration at the values of the other half. Yet you do it to Islam / Arabs / Muslims, because, frankly, you are racist against Muslim people.
An argument that depends on making a blanket statement about a group of people fails for many reasons: categorization (how do you accurately and scientifically select who falls into this grouping and who doesn't?), resolution (how do you account for outliers within this grouping, and how do you determine who might be an outlier?), absolutism (how do you account for the fact that people change?), and due to above, how could you justify making any decisions based on a prejudiced framing?
Racist arguments are completely dependant on fallacy. With a rigid application of rational reasoning, they fall apart. They're illogical.
In what kind of deranged world are we living that people are fighting against the notion that waging war on another country should be a costly decision!?
My, the Overton window has indeed shifted far.
Your philosophy would see the whole world at war.
Counsel: "How do you explain the nanny cam footage of you planting a weapon?"
Robot: "I have encountered an exception and must power off. Shutting down."
> anytime a person in power feels like assassinating a leader or taking out a dissident
I don't really see much of a difference nowadays
People are always worried about getting rid of humans in decision-making. Not that humans are perfect, but because we worry that buggy software will be worse.
Eventually, unfortunately, we will build these systems but it is weak to argue that the technology isn't ready right now and that is why we won't build them. No matter when these systems come on line there will be collateral damage so there will be no right time from a technology standpoint. Anthropic is making that weak argument and that is primarily what I am dismissive of. The argument that needs to be made is that we aren't ready as a society for these weapons. The US government hasn't done the work to prove they can handle them. The US people haven't proven we are ready to understand their ramifications. So, in my view, Anthropic shouldn't be arguing the technology isn't ready, no weapon of war is ever clean and your hands will be dirty no matter how well you craft the knife. Instead Anthropic should be arguing that we aren't ready as a society and that is why they aren't going to support them.
> War is not moral. It may be necessary, but it is never moral.
Nobody who starts a war today acknowledges that they do this for other reasons than "survival", e.g. for stealing various kinds of resources from the attacked.
It has become difficult to distinguish those who truly fight for survival from those who only claim to do this.
The kind of modern wars we're discussing now are often not about survival. Often, the initiator of the war wants dominance rather than survival.
This completely changes the equation. I do pass judgement on those who would wage war to ensure their dominance and access to resources.
However in the case of Iran, who openly calls for the destruction of America and is blatantly developing technology that seriously threaten America and other Middle Eastern states, decisive military action to prevent the threat is important. Don't watch the bully themself and wait for him to confront you, when he is telling you the whole time his intention to destroy you.
Agreed that some countries seek to dominate other regions by force or threat, but you and I are not thinking of the same "groups".
> However in the case of Iran, who openly calls for the destruction of America and is blatantly developing technology that seriously threaten America and other Middle Eastern states, decisive military action to prevent the threat is important. Don't watch the bully themself and wait for him to confront you, when he is telling you the whole time his intention to destroy you.
No, Iran poses no real threat to America, and according to Trump last year suffered a 10+ year setback in their nuclear ambitions. Do you think Trump was lying back then, now, or both?
The US is asserting dominance. Even Trump occasionally says so. Iran mostly poses a danger to their own citizens and, arguably, against Israel when conflict flares up in the region, but not to the US.
By the way, the current situation in Iran is heavily influenced by actions by the UK and the US in the region, back in the 50s. So maybe meddling is not the right course of action?
I guess let the record state that I am deeply morally opposed to automated killing of any kind.
I am sick to my stomach when I really try to put myself in the shoes of the indigenous peoples of Africa who were the first victims of highly automatic weapons, “machine guns” or “Gatling guns”. The asymmetry was barbaric. I do hope that there is a hell, simply that those who made the decision to execute en masse those peoples have a place to rot in internal hellfire.
To even think of modernizing that scene of inhumane depravity with AI is despicable. No, I am deeply opposed to automated killing of any kind.
The “machine gun” has a more complicated history, and the first practical example may have been Gatling’s, or an earlier example used in Europe https://en.wikipedia.org/wiki/Machine_gun
This is a strange take that I didn’t expect to hear. I suppose that the strongest defensive systems do require the most sophisticated offensive systems to defeat, in theory. But there exists asymmetry there as well ($50k drone destroys $1B radar).
My take on weapons development is that there were plenty of mass killing (or mass punishment) devices developed specifically for use by colonial powers against indigenous peoples. This happened alongside weapons development for weapons intended for, as you put them, peer-adversaries.
Revolts happened, and colonial powers needed effective ways to keep indigenous peoples enslaved.
> Putting real lives on the line makes war a costly last resort.
But the AI cat-for-war has left the box for both Iran and the US. Opposing US development of AI for warfare will not suppress US's adversaries from developing the technology.
https://en.wikipedia.org/wiki/Roger_Fisher_(academic)#Preven...
> Fisher [...] suggested implanting the nuclear launch codes in a volunteer. If the President of the United States wanted to activate nuclear weapons, he would be required to kill the volunteer to retrieve the codes.
>> [...] The volunteer would carry with him a big, heavy butcher knife as he accompanied the President. If ever the President wanted to fire nuclear weapons, the only way he could do so would be for him first, with his own hands, to kill one human being. [...]
>> When I suggested this to friends in the Pentagon they said, "My God, that's terrible. Having to kill someone would distort the President's judgment. He might never push the button."
> — Roger Fisher, Bulletin of the Atomic Scientists, March 1981[10]
"If we develop <terrible weapon> we can save so many lives of our soldiers". It always ends up being used to murder civilians.
That's to say nothing of the deaths in a potential US/USSR conflict that goes hot without the Damocles Sword of MAD...
"Based on a detailed investigation of all the facts, and supported by the testimony of the surviving Japanese leaders involved, it is the Survey's opinion that certainly prior to 31 December 1945, and in all probability prior to 1 November 1945, Japan would have surrendered even if the atomic bombs had not been dropped, even if Russia had not entered the war, and even if no invasion had been planned or contemplated."
While this is all speculation, that was at the very least a defensible point of view held by a bunch of Americans shortly after the war.
Regarding firebombing: Hiroshima alone killed probably more civilians than the entire Tokyo firebombing campaign. A firestorm is a terrible thing, but you can still run from a fire even if your whole city burns down; you can't run from a nuke.
So if you measure collateral damage primarily in civilian deaths, firebombing still looks much better (a hypothetical firebombing campaign would have probably killed <40k civilians in Hiroshima instead of 100k, guesstimating from Tokyo numbers).
Edit: I don't think dropping the nuclear bombs was especially ethically questionable compared to the rest of the war, but I feel it is very important to not whitewash that event as valiant effort to save young American conscripts. Regarding it as a slightly selfish weapon demonstration feels much more accurate to me.
Nuclear bombs appear as uniquely horrifying and requiring special justification only in hindsight. Back then, it was just another type of bomb. The thought process behind dropping it was simply "let's hit them as hard as we can until they surrender".
To some of the military leaders, sure. To the scientists and politicians, it wasn't viewed through such a simplistic lens.
I disagree slightly with that take. Decisionmakers knew that those singular bombs were gonna glass an entire city each, and previously almost untouched targets were selected to better show and observe the effect.
If you're at a point where you can afford to slash the primary target (Kyoto) because of nostalgic value to your secretary of war then it becomes difficult to rationalize the whole thing as "normal genuine war effort" and makes the thing look somewhat of an optional choice.
But from my point of view much more questionable decisions were made than the atomic bombings, and hindsight is always 20/20.
I read that this was not the primary motivation; rather, those cities were basically on the top of the "list of industrial centers we didn't get around to bombing conventionally yet, but were going to do next".
Because the Japanese were still fighting, the wartime economy was still trying to produce what it could, and the Japanese government was arming and instructing its civilians for fighting.
Post-surrender, it’s easy to say “we were planning to surrender”, especially to save face and pass the buck to the Americans for what the Japanese government brought on their people by continuing to prosecute the war effort.
I don't know, but there's a lot of evidence this wasn't a factor in the decision to drop the bombs on Japan. The planners for the invasion and the planners for the bombing weren't exactly talking to each other and coordinating the strategy.
They had the bomb and they were going to use it. Everything else was an a posteriori justification.
Now think what will happen with easily deployed AI-powered weapons.
Not so clear cut. Because now sending people to die in distant wars is likely to get a negative reaction at home, this creates some sort of impediment for waging war. Sometimes not enough, but it's not nothing. Sending your boys to die for fuck knows what.
If you're just sending AI powered drones, it reduces the threshold for war tremendously, which in my mind is not "the moral choice".
All of this assuming AI is as good as humans.
Why do people keep falling into traps of anthropomorphize companies like this? What's the point? Either you care about a company in the "for-profit" sense, and then money is all that matters (so clearly OpenAI currently wins there), or you care about pesky things like morality and ethics, and then you should look beyond corporations, because they're not humans, stop treating them as such. Both of them do their best to earn as much as possible, and that's their entire "morality", as they're both for-profit companies,.