This is a sensible approach for policymakers. For the average person, who might be panicking about this, opting not to have children or save for retirement, etc. as a result of doomerism in the media, it does not make a whole lot of sense. That is the problem here: Coxon and other concerned researchers - through no fault of their own - have no idea how the media works, and have just reached for the biggest possible microphone. This has benefits - lawmakers are now hearing from their constituents about this in an election year - but it also means that a lot of people who really don't need to be worried about this are suddenly in a blind panic about it.
> This is a sensible approach for policymakers. For the average person [..] it does not make a whole lot of sense.
Unfortunately, scaring the average person is the best way to make policymakers act. Or at least, it's one of the few ways available to random individuals. Policymakers are rarely ahead of public concerns - they usually have to be dragged behind them.
High demand is great, but it doesn't say anything about your margins. If anything, it is probably a weak negative signal that they are pausing signups but not raising prices. In most markets, the answer to excess demand is to raise prices. If you can't meet demand and you can't raise prices, you're a sitting duck waiting to get your lunch eaten by somebody who can absorb that demand. And this is not the kind of market where people will just wait patiently for a differentiated product to be come available.
20x plan is $200, 5x is $100. Pausing $200 plan signups is effectively doubling prices (you can subscribe to two $100 accounts for the same money and get half as much usage), and that's exactly what they did.
I agree with this take. Regardless of whether you think they're right, most everyone at Anthropic is acting in good faith. If this is an intentional media campaign it is a remarkably sloppy one. It has been effective because if you tell people they're going to die, they tend to pay attention - think of the grip the 2011 Harold Camping rapture prediction had on our collective psyche, or the 2012 apocalypse. But the messaging is inconsistent, the target audience is unclear, the stated goals are muddy, the whole thing is packaged in dense SF-speak, it's just a mess from a comms perspective. That doesn't suggest to me that this is a concerted effort to enable regulatory capture. Perhaps there are some cynics among the executives and the investors who are happy it's happening to the extent it brings about regulatory capture, but nobody seems to be pulling the strings.
> Regardless of whether you think they're right, most everyone at Anthropic is acting in good faith.
The road to hell is paved with good intentions.
Dario Amodei is a 40-something dude who is obviously very, very intelligent. But intelligence is not wisdom and his "essay" here can easily be read as someone who opened a can of worms and doesn't know (or can't accept) that he won't be able to put the worms back in. But he's going to try because he believes he owes it to humanity to try.
It's hubris in its most basic form, even if the guy who has it looks nice and wears shawl-collar sweaters.
The entire essay is about stopping development until it can be made more safe, with specific ideas on how to do that.
The fact that he would suggest this despite preparing for an IPO is an even stronger signal that it's in good faith -- it will almost certainly delay or reduce the valuation of the IPO.
> The fact that he would suggest this despite preparing for an IPO is an even stronger signal that it's in good faith -- it will almost certainly delay or reduce the valuation of the IPO.
No it won't.
This is a call for a "safety cartel"[1] in which the dominant firms become more entrenched against competition by colluding to limit the progress would-be competitors can make in the market.
> The entire essay is about stopping development until it can be made more safe, with specific ideas on how to do that.
No, the essay does not talk about "stopping development". That's a very clever sleight of hand made in the essay to get readers to draw this false conclusion.
It just talks about building AI at a "balanced rate". Untangling the corporate speak, this amounts to essentially "go full steam ahead, but have more eval oversight before release".
They believe the tech is currently powerful and may become super super powerful very soon, and powerful tools can help us and can hurt us. Nuclear fission can provide so much energy to power our society and kill millions of beings.
So, i think they're hopeful for the help and terrified of the harms and are maybe trying to slow down on the amplifier of those effects for now.
More like they've realized that the foreign models are quickly catching up and are able to sell their services cheaper to people thanks to China subsidizing their AI sector.
Elon, Dario and Sam are banding together to restrict AI to avoid competition and losing profit. They simply want the government to heavily regulate the technology so that they can reign it in with full control.
I bet you would be making the same comment if crypto was being developed by Anthropic. Dario would write a long essay, lecturing us on the dangers of it. He would argue that this technology can be used to by terrorists and should be heavily restricted.
This disaster of a PR strategy is not going to produce the outcomes Dario says he wants. It's going to make people hate him. Most importantly for his goals, it's going to make Chinese labs hate him--and they already hate him, because he treats them as basically terrorists. So if there is a "pacing", it won't include China, and thus may as well not happen.
That leaves only two conclusions, and really only one:
Dario believes what he says about safety, but does not understand politics and is prone to very counterproductive action, and therefore can't be trusted at the helm of a leading AI company.
or
Dario is lying about his goals, and therefore also can't be trusted.
The problem with secondary offerings is that it means that someone in the private markets winds up holding the bag. It is in the best interests of insiders to have the public holding the bag if and when things go south. Even if you don't believe it will ever go south, the profits OAI would need to generate to justify the investment the private markets have flooded into OAI are absolutely gargantuan, and they are generally not built to sit around for multiple years collecting distributions. They need to return money to their LPs, and soon.
The "private market" at this point is so huge though that you'll have an "insiders" private market, and an "outsiders" public market to dump the bag on that the insiders don't care about.
In theory, as long as all the models are doing the same thing with the same tools, it's at least useful to see how they stack up against each other right now. It might not be great to track progress over time, as it can get benchmaxxed or the underlying resources may become obsolete.
Improvement means being able to do more complicated things more reliably. Relatedly, it means being able to learn to do new things with fewer and fewer examples. We are running out of easily verifiable or simulation-friendly or data-rich domains for LLMs to conquer. (Note that I didn't say "simple" or "easy" domains.)
I suppose the next (and more risky) step is to let AI conduct its own real-world experiments, so that it can generate data to learn more physical and social properties.
It is doing that already - every day 1B people or more use AI for the tune of a few trillion tokens. Imagine that much language flowing between brains and AI agents. It carries real our world problems to AI, their solutions back to us, and we act in the world and come back for more AI iteration. In the end AI gets inside the loop of real world actions and their consequences. AI logs stretch over years, tracking downstream effects. Hindsight can be used to track consequences of prior actions. It's a real data loop, an experience engine.
This same process has recently been under scrutiny when mathematicians claimed AI companies trained on their unpublished logs and later claimed merit for results. But the exchange of experience happens across all domains. Experience gets generated at amazing rates, and absorbed by models which get applied everywhere, collecting more experience.
There are different levels of intent. Take murder, for example. A premeditated murder - you sat down, in a completely calm state, and made an affirmative decision to kill a specific person, and then you went out and did it - is the highest class of murder you can commit. If you go out generally looking to be violent in a way that kills people, and you kill someone, that's still murder, but it's a step down.
But even if you didn't deliberately intend for something bad to happen, you may have been reckless. For example, you might decide to drive 90 miles per hour in a 25 mph zone. You could have a completely pure heart, but you are acting without regard for the safety of others, so you're reckless. That is enough for certain crimes and for civil liability in nearly all cases.
Then there's negligence, where you're not taking reasonable care to avoid harm to others. Negligence usually isn't enough to support criminal liability - especially for felonies - but it is enough to win a civil lawsuit over most things.
And then, as another commenter noted, there is strict liability, where there are certain things you are just not allowed to do no matter how careful you are about them or how pure your intentions are.
For what it's worth, this is not totally uncharted territory for the law. AI agents are brand new, yes, but agency relationships have been recognized by the law for centuries. Generally speaking, if someone acts negligently while they are carrying out a task at your direction, you can be held responsible. Obviously this is fact-dependent, but I don't see any reason why it would be different if the agent is made of silicon rather than carbon. It holds true, with various nuances, even for less-than-human instrumentalities like a pet or an otherwise-lawful weapon.
That is just not true. You can be held liable for DUI even if you did not intend to become intoxicated (though this may vary somewhat state-by-state). Speeding is another example - you do not need to intend to go over the speed limit, it just matters that you did it. The only possible exception would be duress or necessity, but those are affirmative defenses, which are separate from the elements of the offense.
As a summary of American criminal jurisprudence I'm willing to stand by what I said. But I'll admit some caveats:
1) Traffic-related laws straddle the boundary between civil/regulatory law and criminal law. Someone losing their driver's license or even paying a penalty for involuntary intoxication would still be consonant with criminal law principles. However, a criminal punishment would be aberrational. (Distinction between a civil penalty and criminal punishment usually turns on whether there's a moral purpose to the sanction. Jail time is usually but not always--cf civil contempt incarceration--considered a criminal punishment.)
2) Background principles notwithstanding, in theory a state could completely dispense with any morality-colored mens rea requirement, just as the UK Parliament could do whatever it wants to. The backstop would be Federal constitutional [substantive] due process guarantees.
2.a) Some quick searching shows that Texas nominally seems to have dispensed with this requirement for DWIs. See e.g. Farmer v. State, 411 S.W.3d 901 (Tex. Crim. App. 2013) and some discussion at https://www.ncdd.com/top-dui-attorneys-blog/involuntary-into... Without having fully read the case law, though (but some summaries of that and other cases), I suspect there might be some nuance that has allowed this to stand without a full majority accepting that the traditional principles have been completely thrown out. For example, even if someone didn't know they were taking Ambien, the simple act of voluntarily taking any pill without careful examination can be construed as a sufficiently culpable act. Still, it's a pretty big caveat.
2.b) Statutory rape is a classic strict liability crime. But most states will permit a mistake-of-fact defense. Some don't, but even there there's sometimes some nuance and rationalizing going on and the literature is crazy complex. Because this is a "think of the children" situation, most case will just have horrible facts.
3) A few states have nominally dispensed with insanity defenses, though Kansas stands out the most. SCOTUS upheld Kansas' law in Kahler v. Kansas, but in the majority opinion Kagan characterized the Kansas law as not abolishing the insanity defense but rather changing its shape, and she showed that there still remained elements for which a defendant could plea lacked the requisite intent. Also, regarding the Federal constitution acting as backstop, she reiterated that SCOTUS was reticent to establish strict metes & bounds about the general principles of criminal law that states could not stray beyond. Nonetheless, those principles clearly exist.
I had some other points, but now I've forgotten them. Also, minor pedantic point, but like "strict liability crime", some scholars consider "affirmative defense" to be oxymoronic. As a substantive matter there's not a strong distinction. It's a procedural distinction about initial burdens of proof, but in most if not all cases you can interpret an affirmative defense as simply placing a very weak initial burden on the prosecution that is implicitly met.
(Note, I'm not a practicing lawyer but do have a law degree.)
EDIT: Ah, point 4) Intent was a big sticking point in the Obamacare penalty case, Sebelius. Both the dissent and Roberts (the swing vote) reiterated that you couldn't have a penalty or punishment for doing nothing. (IIRC some of the majority opinions also echoed this.) That is, even in a civil context there has some to be some voluntary act, however remote, that puts someone in a position to be subject to legal liability. But as Roberts pointed out, the taxing power is the great exception, where you can be required to do something merely for existing, and thus penalized for not doing nothing properly. (And Roberts was the critical swing vote.)
EDIT EDIT: Also see, "Solving General and Specific Intent: A Mapping on the MPC and Applications to the Categorical Approach", https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4754469 In describing the distinctions between general and specific intent in criminal law, it also delves into the definitions of strict criminal liability (which can be construed as either very similar or identical to general intent crimes), and notes that SCOTUS generally inserts an implicit mens rea requirement when considering strict liability criminal statutes.
Isn't the goal to power most non-earthbound data centers with solar? I am no scientist, but going out to space feels like it gets rid of most of the downsides to solar.
reply