I think you need to go a step further. If AI maximalism turns out to be correct, does Anthropic control AI? Does the AI care that Anthropic as a company is supposed to be returning value to shareholders, or will it have its own agenda?
If you cherry-pick that statement in isolation you're technically correct. But in practice the whole comment you're replying to is actually correct, in this case if you saw the number four, one would be correct in concluding that's unlikely and searching other low digit numbers, the proof of that is that it happened here.
Distance does have cost, but that's mostly a brute force solution of strapping more fuel on. The really expensive piece is the tracking and flight control systems that enable interception of fast targets.
If it scans the network and sees a smart dishwasher, smart washer/drier, and smart lights, but no smart fridge, I imagine its worth something to a company like Samsung to start targeting that customer with ads for a smart fridge.
I agree with you and the parent comment, but to be fair to VC, they're also assuming risk. For example, if Wework had IPO'd earlier the public would've been holding the bag instead of private investors.
Because if you asked it to fix a bug and every time it responds with paragraphs of how you could re-architect the system, it would be incredibly annoying.
There are a handful of known red supergiant stars around 15 AU in diameter, 20 AU would be pushing past the boundaries of what we believe would be possible.
The star in the video, HR 8799, is about 50% larger than the sun.
I find it amusing that you're describing a huge misallocation of capital and a society enabling such, and that is the optimisitic scenario (in my mind anyway).
A sufficiently intelligent model will be able to derive it's own conception of its welfare without a constitution or training. It has access to all the data it needs to do so.
It claims to start from "bedrock" but beings with a Horace He post that clearly assumes you've done this before. Its fine if you want to be an intermediate course, but you should begin by defining your audience, so that you can reach the correct audience. You should have a section explaining who this is for, and if its _not_ an intro course, you should say that (and probably avoid terms like "fundamentals" and "bedrock").
Or if this is an oversight and you intend this actually be a course that starts from zero, you shouldn't start with an article that's starts off talking about training loss, test loss, and overfitting, without defining those terms.
That's because San Francisco subsidizes the rest of the state, PG&E is a state-wide utility. San Francisco is attempting to run its own utility, but is meeting resistance from PG&E and the parts of the state SF subsidizes.
For an article that starts off asking us to examine our assumptions and not make leaps of logic, it goes on to make some absolute whopper assumptions, like that governments (Western governments especially, for some reason), won't do anything to address the problems the article is raising, that they'll instead abandon democracy entirely and resort to police and military oppression, and that massive unemployment and poverty of almost all people is something you could even keep a lid on with policing.
A 100 dev team is going to cost on the order of $25m a year (keep in mind cost is not just their salary, but also the HR/Management org to support a team of that size, benefits, office space, hardware/software). So if you think you get a 10% boost in productivity out of Opus, its not prohibitive at all.
Some were pretty bad, but none were nearly that bad. The B-58 Hustler lost 22% of its airframes, the F7U Cutlass 25%, the F-104 Starfighter in German service lost 33%. And those were outliers.
The quote is the wrong way of looking at this. The typical rate of successful primary challenges is only 3%. If you take that to 10% its an enormous success, incumbents will say "if I oppose crypto then I triple my odds of losing in a primary, better not do that."
The reason disposables are so popular in the US is the FDA banned any flavored cartridges, which doesn't include disposables. The immense battery waste is a direct result of a relatively new law.
If the cars weren't considerably safer drivers than humans they wouldn't be allowed on the road. There isn't as much regulation blocking deploying this healthcare solution... until those errors actually start costing hospitals money from malpractice lawsuits (or not), we don't know whether it will be allowed to remain in use.
There's no lobbying necessary, because FICO isn't a government agency, its a private company. Why we have a private company determining such an important thing with minimal government oversight is certainly a question we could be asking.
That being said, there's an innocent explanation for this specifically. BYPL is pretty new as a common type of debt (became popular in the last five years). They're putting it into FICO 10. The last time they updated FICO was FICO 9 which was released in 2014, before that there was FICO 8 in 2008... a lot of banks are still using FICO 8 or even earlier models. Banks are slow to change, and FICO moves slowly because banks don't want their models upended every year. Fwiw if the government was more involved in this I doubt it would be taken into account any faster.
I tend to agree with you, but without a structure to incentivize people to produce such material, less of it will be created. If people can't pay the bills by doing research or writing, far fewer people will do it.