Comment by neya
8 hours ago
>Dario on the other hand seems to have an integrity that's particularly rare in this era.
Anthropic actually partnered up with Palantir. They are not the saints you think they are, either.
We should stop worshipping people and companies and stop putting them on pedestals. Just because one party is at fault, doesn't mean the other is automatically innocent. These are all for-profit companies at play here.
https://investors.palantir.com/news-details/2024/Anthropic-a...
FWIW he gives his ethical reasoning on his website:
> Broadly, I am supportive of arming democracies with the tools needed to defeat autocracies in the age of AI—I simply don’t think there is any other way. But we cannot ignore the potential for abuse of these technologies by democratic governments themselves. Democracies normally have safeguards that prevent their military and intelligence apparatus from being turned inwards against their own population, but because AI tools require so few people to operate, there is potential for them to circumvent these safeguards and the norms that support them. It is also worth noting that some of these safeguards are already gradually eroding in some democracies. Thus, we should arm democracies with AI, but we should do so carefully and within limits: they are the immune system we need to fight autocracies, but like the immune system, there is some risk of them turning on us and becoming a threat themselves.
Basically, he's afraid that not arming the government with AI puts it at a disadvantage vs. other governments he trusts less. Plus, if Anthropic is in the loop that gives them the chance to steer the direction of things a bit (what they were kicked out for doing).
It's not the purest ethical argument, but I also would not say that there is a clearly correct answer.
Basically he's asking everyone to trust him that he won't cross the line himself. Whatever argument he makes for democracies applies to him as well, and he's not somehow above it. That's the flaw in his argument.
Brutally honest, to me it just sounds like a very elaborate way to say "trust me, bro"
I would agree if not for the fact that they just let a $200M contract slip through over it. You could argue it's "safety theater" in itself but that seems like a risky gambit especially with this administration. I definitely trust Anthropic more than OpenAI. In fact I'd go as far as to say it's probably pretty imperative that Anthropic stays a frontrunner in this race and doesn't leave the field exclusively to OAI (and maybe Google which is just as bad). That doesn't mean I'm exactly happy with Anthropic's comments like "mass surveillance bad but only for the US". But Anthropic at least regularly asks questions about the direction of AI development. I haven't seen the other frontier model companies do any such thing.
2 replies →
If you look at his comments about Palantir and their proposed safeguards, it's clear it's a case of "if you are dining with the Devil, you'd better bring a very long spoon"
These comments were after the deal had soured. Not before. If it was a case of such morality, the partnership with Palantir would have never happened in the first place.
The contract was explicit - it was for defence purposes with a company known for spying activities. So, obviously spying is involved and they weren't just going to generate cat videos with it.
Again, nobody is innocent here.
I've heard Palantir is essentially the only federal cloud vendor with this administration for secure services. By "partnered up with Palantir", do you mean they provided their models to the government? Or something more?
From the title of the link enclosed:
"Anthropic and Palantir Partner to Bring Claude AI Models to AWS for U.S. Government Intelligence and Defense Operations"
Keywords: "Government Intelligence"
> Anthropic actually partnered up with Palantir. They are not the saints you think they are, either.
And now you've got people on here saying, well actually Palantir ain't so bad, you see! They're just one tool in the chain, basically just boring data integration, like IBM!
The mental gymnastics is difficult to keep up with.
If you actually read the memo they've clearly put in strict terms with Palantir and rejected many of the false "safeguards" offered by the company