hypersoar

Privacy, mainly. I don't see how it would make economic sense. $10k will buy an awful lot of GLM-5.3 tokens.

I actually think a desktop makes sense if you want to run local LLMs. I have an M4 Max laptop left to me when I got laid off. It can run good-sized LLMs, but the system heats up, the fans blast off, and it rips through the battery in under an hour. I'd rather have a desktop set up as a private inference server that I could connect to remotely.

I would add: ban all advertising of sports betting.

You...you literally elided over the exact phrase I included in anticipation of this objection.

That is not because Theranos did not commit such fraud against patients; Balwani was convicted on those charges. It's just the prosecutors were unable to prove that Holmes was culpable.

Yes, obviously, that's why terms have been extended so far. My point is that, for the same reason, existing legal structures make this much harder to undo than to do.

Holmes and Theranos leadership did more than defraud investors. They knowingly ran unreliable tests on actual patients. People were told they were miscarrying when they weren't, or that they had HIV when they didn't. And they drove an employee, Ian Gibbons, to suicide, then threatened his widow with a lawsuit when she talked to reporters about it.

I wish, but shortening existing copyright terms would be legally very thorny. It would run into Takings Clause challenges and violate treaties the US has signed.

Other HP fanfics aren't foundational texts for a worryingly influential pseudo-philosophical movement.

While the courts have allowed the influence of money in politics to go much too far, I doubt that the idea of rich people and businesses using their wealth to influence politicians is all that alien to non-Americans.

Also, bribery isn't really "super illegal" in America. Cases like McDonnell vs. US make it so that as long as you don't go "Thank you for the bag of money, which I hereby accept in exchange for granting you, in my official capacity, that government contract you wanted, which you would not have gotten otherwise," you're pretty much in the clear.

The incompleteness theorem says that there are statements which can be neither proven true nor false in a given axiomatic system. If there is a proof to write in lean, then the statement is already outside the bounds of incompleteness.

What the OpenAI agents did is basically how my class in middle school improvised a chat room. We renamed files in a chosen directory. It made me think that some of the reporting about the technical ingenuity shown in that incident was a bit overblown, at least on that point. I mean, we were able to figure it out as normal 12-year-olds.

Parent's link about Stevens was separate from the quote, which is from Warren Burger. And Burger was hardly a lefty. He was appointed by Nixon specifically to begin the work of dismantling the achievements of the Warren court.

The Trisolarians are supposedly at Alpha Centauri, a trinary star that isn't actually that chaotic. It has two sun-like stars orbiting each other quite close, and then a red dwarf, Proxima Centauri, orbiting that pair really really far away.

I'm not necessarily opposed to any AI assistance in the creative process, though it's a complicated question. But there's investment out there in generating content purely for indulgence, and to that I say no fucking thank you. "Here's what it would've sounded like if Rachmaninoff had written a 5th piano concerto!" He died 80 years ago. I'll just keep listening to No. 2 and otherwise move on with my life, like a healthy human being.

Whether or not the AI can "fool" me is totally irrelevant to my point. Yes, one-shot prompts will be able to make content I can't distinguish from something made by a human, if they can't do so already. Some of it will probably sneak its way in to my media diet (though the fact that most of the music I listen to was composed over 100 years ago makes it less likely for me than most). But if I were halfway through a novel that was absolutely blowing my mind when I learned it was created by a publisher punching "write me a book that'll sell good" into ChatGPT 8, I would instantly lose all interest. I would also have unbounded fury at said publisher for dishonestly marketing such a book to me. The point of art is not just to indulge my own sensibilities. It is to connect me to other people.

I mentioned this in another comment thread, recently, but I value art, including music, for the fact that another human being(s) created it to express something. I have zero interest in it if it's AI-generated. The thought of endless content created from the void for me to gorge on like oreos fills me with nothing but despair.

I dropped out of a math Ph.D. in 2018, and I'm increasingly glad that I'm not in math research, anymore. While it's cool that we can get these results, I don't think that I'd enjoy being a post-AI mathematician.

>? Especially in art, the anti-AI hate mostly reduces to arguing that art is the process not the outcome.

That is not my objection to AI art at all. The value of art for me is that somebody has an idea to express and they're trying to communicate it to me. That's totally absent from AI-generated material. I'm not categorically opposed to AI, but my interest in automatically generated content tailored to indulge my personal taste is absolutely zero.

I was just trying that yesterday on my M4 Max with the 6bit quant. It started at 40 tps but dropped to 10 once the context loaded up.

The way I've seen it work is that there's a touch screen in the elevator lobby. You tap a floor, and then screen will say "Car C" (along with an audible message). The problem is that then the system can't reassign you to a different car after that. You'll typically have a bunch of different people going to a bunch of different floors, and there would be no reliable way to communicate to each individual their new assignment. Each car will also typically display which floors it's going to when it arrives. I suppose that, rather than assigning a car immediately, you could have passengers check cars as they arrive to see if they're going to their desired floor. Then the system could do reassignment, but people would be scrambling to check every car to see if it's the right one for them, and that's assuming that they understood the system. It would be way too chaotic in practice.

Federal judicial opinions are readily available online. The circuit courts and Supreme Court--i.e., the courts which actually establish precedent--post all of their opinions on their own websites. If your concern is knowing what courts (at least federal courts, where PACER is used) are saying so that you can follow the law, it's all out there for free. What PACER hosts is the rest of the case records: exhibits, briefs, motions, etc.

The case which originated the "yelling fire in a crowded theater" analogy, Schenk v. US, was overturned more than half a century ago. And in that case, the proverbial fire-yellers were socialists distributing pamphlets.

This is a tangent to this post, but...

I happen to have a cat named Kiki who looks rather like the mascot for this project. Her health is failing, now. I just spent the night on my living room floor next to her. I'll, likely have to put her down, today.

I might use this project to make a memorial page for her.

https://ibb.co/7dRCnWrp https://ibb.co/1GWwDKLY

The article says that California "will require identity verification at the operating system level starting in January 2027", but that isn't true. The bill [0] only requires that operating systems collect age information on account setup and then provide which of four buckets the age falls into. It's not requiring any kind of identity verification by the OS. It's just a box you fill in when you set up the computer.

I think that this is actually a reasonable approach. It minimizes the information shared and doesn't create any identity tracking regime.

[0] https://media.reclaimthenet.org/docs/california-ab-1043-digi...

I can find no evidence that California ever tried "banning high school calculus". The chapter in the much-maligned mathematics framework on high school [0] makes no such proposal, and indeed suggests consolidating the prerequisite classes to make it easier to reach calculus without acceleration in middle school:

> An alternative to eighth-grade acceleration would be to adjust the high school curriculum instead, eliminating redundancies in the content of current courses, so that students do not need four courses before Calculus. As enacted, Algebra II tends to repeat a significant amount of the content of Algebra I, and Precalculus repeats content from Algebra II. While recognizing that some repetition of content has value, further analysis should be conducted to evaluate how high school course pathways may be redesigned to create more streamlined pathways that allow students to take three years of middle school foundations and still reach advanced mathematics courses such as calculus.

Nor can I find any evidence that they "reject the idea that some kids are more talented at somethings than other kids". Instead, their FAQ [1] includes:

> All students deserve powerful mathematics instruction. High-level mathematics achievement is not dependent on rare natural gifts, but rather can be cultivated.

> All students, regardless of background, language of origin, learning differences, or foundational knowledge are capable and deserving of depth of understanding and engagement in rich mathematics tasks.

This is not remotely the same as the silly framing of "if you can compute a derivative by 12th grade, it's due to racial discrimination". It's about not giving up on students who are undeserved by mathematics education as it is currently constituted.

I myself have mixed feelings on "de-tracking" mathematics courses. I benefited from accelerated math classes and would have been bored to tears if forced to take classes at the standard pace. But I also understand that accelerated classes have tended to allocate more resources to students who are already succeeding. It's a thorny problem. But this comment adopts the framing of right-wing propaganda rather than the actual contents of the framework.

[0] https://www.cde.ca.gov/ci/ma/cf/documents/mathframeworkch8.p... [1] https://www.cde.ca.gov/ci/ma/cf/mathfwfaqs.asp

I attended Google I/O in 2013 and was given a Chromebook Pixel, their $1300 laptop. The hardware was very, very nice, and I quite enjoyed using it for a while. One day, I dropped it and damaged the screen well outside of its warranty period. "Oh no," I thought. "This is probably going to be pretty expensive to fix." So, bracing for the damage, I called up Google and told them what had happened. They replied that there was no fixing it. They would replace the laptops under the warranty, but there was no repairing to be done. I was welcome to call around and ask local repair shops if they could do it. That went nowhere, of course.

I've been pretty skeptical of Google laptops ever since.

The author (who is also the submitter; it seems nearly all his submissions are his own blog posts) is not a physicist, so it's hard for me to take seriously his sweeping dismissal of the field. Then I see he links to his own Revolutionary Theory, and it starts to look like outright crankary.

That choice is doubly false. On the one hand, there was a diplomatic option. It was working until Trump decided to kill it. On the other, it's insane to think that you can bomb a large, industrialized country of 90 million people out of the ability to make nuclear weapons short of wiping them out of existence.

I can believe that such an atmosphere exists there. I can't believe that it will stay. It will be squeezed out by the drive for profit in time.