He has to know that this would leak and it makes him look really bad. This is going to be a meaningful, unforced error.
....why does this make him look bad? That he called out the obvious thing that everyone knows?
Dario Amodei calls OpenAI’s messaging around military deal ‘straight up lies’
81–90 of 461 posts
Re: Dario Amodei calls OpenAI’s messaging around military deal ‘straight up lies’
#82It’s all just theatre. These companies will either give in or die off and be replaced by those who offer more freedom of use. It’s capitalism and while it’s not always pretty, it’s how these things go. Choosing to take what you believe as the moral high ground is noble but it does not put your company ahead of the ball in the long term because there are always those who will use that as an advantage to step on their…
Capitalism needs laws and regulation in order to not turn itself into feudalism. It isn't naivety or idealism to enforce fair markets and consumer protection. In my opinion it's existential.
Re: Dario Amodei calls OpenAI’s messaging around military deal ‘straight up lies’
#83Honest question: why do people automatically equate "fully autonomous weapons" to something like killer robot? My immediate reaction is that even the best-in-class rapid-fire gun has a hard time identifying and tracking drones. So, we'd need AI to do better tracking, which leads to a fully autonomous weapon. And I really don't get why that's a bad thing. Of course, a company should have freedom to choose not to do bu…
We have traditional autonomous weapons (and counter-defense). They operate on millisecond or faster timescales with existing RF sensors. They are not and will not be using LLMs or other transformers. Maybe ChatGPT will update some realtime Ada code; they formally verify some of that stuff so maybe that won't be terrifyingly dangerous.
Where autonomous transformer-based munitions will be used are basically "here is a photo of a face, find and kill this human" and loitering munitions will take their time analyzing video and then decide to identify and attack a target on their own.
EDIT: Or worse: "identify suspicious humans and kill them"
Re: Dario Amodei calls OpenAI’s messaging around military deal ‘straight up lies’
#84Leaving autonomous weapons aside, how does Anthropic justifies that they signed up with surveillance company Palantir and now raising concerns for same surveillance with DoD? It doesn't match.
[flagged]
"The law" is the contract. The Pentagon agreed to terms of service. The law is not on the Pentagon's side. The contract did not change; what changed is the Pentagon breaking the contract.
Perhaps you think the law shouldn't allow such a contract; that's a valid position. But that's not what the law currently says.
Re: Dario Amodei calls OpenAI’s messaging around military deal ‘straight up lies’
#85Like others have already mentioned: I think Anthropic's relationship with Palantir undermines Amodei's narrative here. It actually feels like Dario is playing Sam's game better than Sam is. Those who know better please correct me. My current understanding of Palantir (and other surveillance tech companies like Peregrine) is: 1. They facilitate the sale of data to law enforcement, enabling the government to circumvent…
We don’t know if Palantir is using claude for those uses. Though anthropic would not know for sure either.
I do agree with your point that Amodei is playing a game though. Whether he’s winning the bigger picture or not it’s unclear. His red lines are already so watered out, like how domestic surveillance is not ok, but international? totally fine.
Re: Dario Amodei calls OpenAI’s messaging around military deal ‘straight up lies’
#86Earlier quoted context omitted.
[flagged]
If they're doing it against the terms of service (and publicly so), I can't pin that one on Anthropic. They've done lots wrong and maybe they shouldn't have gotten in bed with the military to begin with, but this illegal war is not theirs. It rests squarely with the President who declared it. (And with the military officers who are going along with it despite the violation of international law.)
I don't think any AI company should get in bed with the military. That being said, if the terms of service have been violated, the account should be canceled.
Re: Dario Amodei calls OpenAI’s messaging around military deal ‘straight up lies’
#87Like others have already mentioned: I think Anthropic's relationship with Palantir undermines Amodei's narrative here. It actually feels like Dario is playing Sam's game better than Sam is. Those who know better please correct me. My current understanding of Palantir (and other surveillance tech companies like Peregrine) is: 1. They facilitate the sale of data to law enforcement, enabling the government to circumvent…
We don’t know if Palantir is using claude for those uses. Though anthropic would not know for sure either. I do agree with your point that Amodei is playing a game though. Whether he’s winning the bigger picture or not it’s unclear. His red lines are already so watered out, like how domestic surveillance is not ok, but international? totally fine.
That's true. With the risks of LLMs applied to surveillance though, I think it's a "Caesar's wife must be above suspicion" moment. Association is guilt unless proven otherwise.
Re: Dario Amodei calls OpenAI’s messaging around military deal ‘straight up lies’
#88Honest question: why do people automatically equate "fully autonomous weapons" to something like killer robot? My immediate reaction is that even the best-in-class rapid-fire gun has a hard time identifying and tracking drones. So, we'd need AI to do better tracking, which leads to a fully autonomous weapon. And I really don't get why that's a bad thing. Of course, a company should have freedom to choose not to do bu…
What you are describing would be "partially autonomous." Per Dario Amodei's original statement here: https://www.anthropic.com/news/statement-department-of-war he had no issue with that. "Fully autonomous" specifically means that the AI chooses a target and engages without any human intervention at all. If the human selects or approves a target, and the weapon then automates tracking and engagement, that's still only partially autonomous.
Re: Dario Amodei calls OpenAI’s messaging around military deal ‘straight up lies’
#89Honest question: why do people automatically equate "fully autonomous weapons" to something like killer robot? My immediate reaction is that even the best-in-class rapid-fire gun has a hard time identifying and tracking drones. So, we'd need AI to do better tracking, which leads to a fully autonomous weapon. And I really don't get why that's a bad thing. Of course, a company should have freedom to choose not to do bu…
I’m not sure that “killer robot” is the actual concern outside of media hyperbole. I’m imagining a loitering munition-type drone that has some kind of targeting package loaded into it with different parameters describing what it should seek and destroy. Instead of waiting for intelligence and using human command to put the munition on target, it hangs out and then engages when it’s certain enough that it’s found something valid.
In a world where LLMs produce very convincing but subtly wrong output, this makes me uncomfortable. I get that warfare without AI is in the past now, but war and rules of engagement and AI output etc etc etc all seem fuzzy enough that this is not yet a good call even if you agree with the end goals.
Re: Dario Amodei calls OpenAI’s messaging around military deal ‘straight up lies’
#90Honest question: why do people automatically equate "fully autonomous weapons" to something like killer robot? My immediate reaction is that even the best-in-class rapid-fire gun has a hard time identifying and tracking drones. So, we'd need AI to do better tracking, which leads to a fully autonomous weapon. And I really don't get why that's a bad thing. Of course, a company should have freedom to choose not to do bu…
I’m not sure that “killer robot” is the actual concern outside of media hyperbole. I’m imagining a loitering munition-type drone that has some kind of targeting package loaded into it with different parameters describing what it should seek and destroy. Instead of waiting for intelligence and using human command to put the munition on target, it hangs out and then engages when it’s certain enough that it’s found some…
> I’m imagining a loitering munition-type drone that has some kind of targeting package loaded into it with different parameters describing what it should seek and destroy. Instead of waiting for intelligence and using human command to put the munition on target, it hangs out and then engages when it’s certain enough that it’s found something valid.
I'm sorry, you've just literally described a "killer robot" in more words.