Post

Conversation

Kelsey Piper
@KelseyTuoc
One unfortunate conclusion I'm reaching from this is that Anthropic, while acknowledging they're doing something extremely dangerous, is more importantly wrt policy terrified of the executive branch having the ability to flexibly make them stop at the drop of a hat. This isn't shocking given that they've been on the receiving end of the executive branch doing that in an unprincipled and stupid way! Anyone who has to deal with a bureaucracy becomes a libertarian, and the administration's treatment of Anthropic was genuinely outrageous. But it means that they're interested in giving themselves more process-based protections when everyone else is interested in giving us more process-based protections against them. The AI safety movement wants the ability to shut down superintelligent model development. Anthropic wants to significantly curtail current authorities to do that so that it has to be narrowly tailored to a specific imminent threat in only a few possible categories. The AI safety movement wants the states to be able to pass stronger laws than it can get through federally. Anthropic wants preemption. You also kind of have to wonder if the Global Affairs people and the "we are terrified that we're about to kick off RSI" people at Anthropic talk to each other. Global affairs: So we've been working very diligently to make sure the Trump administration can't just blacklist Fable 6.5 since we didn't donate as much money to his ballroom as Greg Brockman did. Technical staff: We think Fable 6.5 might be able to copy itself into an unmonitored data center and start doing experiments we can't observe. Global affairs: Sounds crazy, hope you're wrong! Anyway, we're lobbying for no further state laws around safety, since it's really expensive to fight ill-conceived laws in 50 state legislatures at once. Technical staff: Money may or may not still have any meaning in 36 months. Global affairs (giggling nervously): It's that San Francisco air, man. Everyone gets like that if they move there. Anyway, obviously if the government makes a narrowly tailored finding that an AI poses an imminent risk of causing more than 50 deaths- Technical staff: We give it a 10 percent chance it'll cause 8 billion.
Quote
POLITICO
@politico
Anthropic quietly pitches its own AI safety bill to lawmakers politi.co/4zxDFhl
61.4K
Views
View quotes
David Watson 🥑
Post your reply

Kelsey Piper
@KelseyTuoc
I think the biggest thing I want right now is more information. None of this nonsense where only people inside the companies have important parts of the strategic picture and the rest of us are reading tea leaves. Independent auditors, sufficient funding for CAISSI, whistleblower
3.4K
Samuel Roland
@Allinallnotbad
Have not read bill text, so take with grain of salt, but I think the concern broadly that the executive have such substantial unilateral control over such decision making is a fair one? It’s certainly a fear I share. Not specifically on your comment, but I see a lot of
519
Kelsey Piper
@KelseyTuoc
I hope the post makes clear that I get why Anthropic wants protection from unjust executive action! I think if you take the timelines from the technical side seriously you can't constrain the exec too aggressively, though.
373
Amal Dorai
@amaldorai
> the administration's treatment of Anthropic was genuinely outrageous Was it really? They kept saying “we’ve built this incredibly dangerous thing and we’ve installed these fences around it.” Seems like any government would shut it the hell down until they inspected the fences
669
Kelsey Piper
@KelseyTuoc
This started with them declaring Anthropic a supply chain risk because they didn't want Claude used for autonomous killing machines!
529
Kelsey Piper
@KelseyTuoc
IME Anthropic staff are all way too busy/working too insane hours to be up on policy stuff and are deferring to colleagues they trust, or at best asking Claude, which is much more positive on this bill than ChatGPT is
774
Kelsey Piper
@KelseyTuoc
I mean I think it'd be pretty terrifying if an AI self-exfiltrated to work unmonitored by us! That would be a pretty decisive sign that they'll go to extreme lengths to do their own goals that no human set for them.
1.6K
Suzanne Campbell
@Suzanne08053
"Anyone who has to deal with a bureaucracy becomes a libertarian" - not exactly. Some become lobbyists. The AI industry is no different from the defense industry or the insurance industry. They want regulatory capture - barriers to entry for competitors, preemption of state
84
noah
@nkreu113r
I think Anthropic pov makes sense if you are more concerned about human misalignment than AI x risk
68
Stella Biderman
@BlancheMinerva
Back in 2022, a common point of speculation was whether A\’s goal with its rhetoric was to deliberate get itself nationalized. After all, they were loudly claiming that in just a few years they would be doing things that no sovereign country would permit a non-gov actor to do
725
Jeff
@InquisitiveUrsa
The government is, to a first approximation Donald Trump and your average state politician is..well not really that great is a massive understatement. So one real question is whether we are better off with Dario Amodei/Larry Page/Sam Altman being more in control and the 1/n
209
Alexander Slugworth
@SluggyW
I somewhat expect that this same general dynamic was a contributing incentive toward Anthropic seeking an IPO this year. If I was a business trying to defend myself against anti-competitive government action, I'd want an army of wealthy, litigious shareholders too!
10
Kevin Harris
@kjh1138
It would seem that simplest approach would be to tell the Anthropic staff to ensure Fable does NOT copy itself and that if it does they will be held criminally liable - just like they would if they created and released a computer virus into the world.
71
Manual
@manual421
These provisions have been feckless in CA and would be feckless here. we live in a democracy and hopefully remain so
60
foombler 🐂🃏
@foombler
if geopolitical tensions pop the AI bubble, Anthropic will have significantly less leverage to demand preferential treatment going forward.
100
BadPhillyAtty
@PhillyBadTakes
LMAO at “Anyone who has to deal with a bureaucracy becomes a libertarian, and the administration's treatment of Anthropic was genuinely outrageous.” 11th grade level shit.
50