cvoss

It's an interesting bind, not at all unlike how some efforts to address racial discrimination end up being themselves racially discriminatory. So it's like, ok, is it ok sometimes or never?

I'd be fine if you argued that Title VII is unconstitutional for this reason. I like Title VII, but I see why it might be argued against on these grounds.

I get that "no religion" is a "religion" for the purposes of not discriminating based on which religion someone is. And if you want to distinguish humanism from "no religion", that's fine. But humanism doesn't bind anybody to believe anything. It's merely a descriptor of a set of beliefs. That's wholly unlike the major and minor religions of the world. If you're a humanist but are ok with AI, nobody and no thing has the authority to tell you you are wrong. Some people may have opinions to share to persuade you, but there is no authoritative answer from a person, text, tradition, or deity. That's the difference.

Courts don't like to wade into religious matters, precisely because of their deference to religious authorities, owing to the 1A. If your church declares you a "minister" be because you help lead kids' Sunday school lessons, SCOTUS is not going to debate that; they just take it as given.

But if you go to court and say, "well I'm a minister in my religion because I said so," that's not going to fly. You're free to exercise religion, not make things up and pretend like your private opinions are religiously imposed on you.

For exactly this reason, there are plenty of people from traditional religions who claim exemptions to things that I think are bad faith uses of Title VII.

His beliefs are not currently protected. He filed a Title VII request, which was ultimately granted by his employer voluntarily. Because no court ruled that the employer had to grant the request, it is an open question whether Title VII protects them.

I would be shocked if, had the university declined, the courts ultimately ruled that it was protected. Congress is free to write laws protecting people against discrimination for various beliefs. But they law they have on the books now shouldn't be stretched to cover this.

It's worth understanding why sincerely held religious beliefs are such a big deal and in a different league from "strong personal convictions". It's because people who hold them generally are under and bound by two authorities: their state law and their church law (which might take many forms). If a person is in a position where they simply cannot abide both laws simultaneously, the state must yield. Otherwise the state is no protector of freedom.

A humanist may have strong, sincerely held beliefs, but he is not under any authority that compels those beliefs, so they aren't "religious". He is permitted by all his authorities to use AI. He just doesn't want to out of a private belief that it is somehow "wrong".

If your employer orders you to do something you believe is wrong, but you are not required to believe it is wrong by a higher authority, then your employment should terminate. Same as if you just didn't want to do your job, your employment would terminate. Title VII just isn't about that stuff.

I would bet no. Because the original proof is so distributed, with parts of it being very old, and with parts of it being, as a first attempt, not the most concise or lucid route to the result, we were rapidly approaching a world where the empirical knowledge of how/why the result holds was literally dying.

The article refers to its use of the "first-generation proof". It would seem this proof is too large to be truly known by one person, and the group of people who "collectively knew" it grew old and started to die. Since its original formulation, mathematicians began to compile, consolidate, and revise the proof into a 2nd generation digested form (estimated to be 5000 pages instead of tens of thousands). But that effort was still so monumental in its own right that a 3rd gen proof project began to digest the 2nd gen proof before the 2nd gen proof was even finished. I believe both the 2nd and 3rd gen projects are still ongoing.

It will be interesting to see how small the proof gets once we finish learning how to obtain these results "efficiently" with the benefit of hindsight.

Indeed, this formalization effort spotted a flawed argument in one of the original publications and worked around it to repair the result.

It's like its simulation model is confined to the renderable area. At the cost of increased computation, it would not be hard to generalize the technique to rendering a sphere centered around the POV to keep some of those artifacts at bay; however, the fundamental limitation of "object permanence" remains if objects can evolve chaotically as soon as they are blocked from view by another object. For that to be made stable, you really need a whole 3D model of the scene, and the cost increase there might be huge.

> While it can maintain coherent environments over short sessions, visual details and object permanence can still shift or drift over time—especially as a simulation runs longer, after multiple complex prompt edits, or when navigating throughout the scene. Complex physical interactions, real-time audio accuracy, rendering legible text, spatial accuracy, and long-term consistency are active areas of ongoing research.

How does open weights address the issue of nation states injecting covert propaganda into another nation's public discourse?

"Pre-print" implies it's headed to be "printed" by a publisher. That is, the paper has already been accepted for publication by a peer-reviewed publisher, and it's just being posted early for wider and faster dissemination or to stake a claim of priority.

Otherwise, the document is a self-published manuscript, which doesn't carry the authority implied by "publication" or even "pre-print".

> plural inanimate abstract words ... "feminine"

A relic of gendered nouns in English can still be seen in how we treat concepts like countries, churches, schools, organizations, languages, and complex machines (especially ships). While usually people ascribe the pronoun "it" to these, occasionally they use a gendered pronoun, and if so, it's always "she", never "he". Two churches or organizations or even ideas might be "sisters" or in a "mother/daughter" relationship. The common theme through all these examples is plurality and abstractness.

> You have it all wrong.

> Apple Reference Image is not an id system

GP does not have it all wrong. A company desiring you to prove your identity often asks for a photograph of your government ID. Now that this is easily faked, it is reasonable to expect that the company will ask for a verifiably authentic photograph of your government ID.

That the Apple Reference Image itself is not traceable to the device/user is beside the point in this use case.

The article abridged the history too much. It was first imported as an exhibition of plants from Japan in 1876. It's usefulness as a ground cover in the US was understood later, and in the 1930s (Dust Bowl times) the USDA advocated farmers intentionally plant it as a way to hold down the soil. After that, you might say, it far exceeded its usefulness as a ground cover.

Common sense cannot be trusted on matters like these. Common sense is calibrated for reasoning over matters encountered in daily life. The further away we get from that, into more and more exotic phenomena, the less common sense can apply. Black holes are very far from the domain of common sense.

It's one thing to argue, even vehemently, that this should not be a felony (either as a matter of fact in this case, or as a matter of just law). But granting that it is or is liable to be prosecuted as such, I strongly recommend against giving advice on how to commit a felony and get away with it.

> Neither the prestigious degree nor the good proficiency in algorithmic problem-solving guarantee a career success. They are factors that make statistics work in the favor of the candidate.

I think the article takes a more nuanced view than you suggest. And the existence of counterexamples to a "for the most part" claim are not counterexamples at all.

> knowing full well that [Delilah] will betray [Samson]

It is not clear from the text that Samson is aware of the presence of Philistines the first three times. We are told they are lying in ambush in an inner chamber. We are not told that they make themselves known, attempt to subdue Samson, or interact with him in any way. Instead, we are told that Delilah announces their imminent presence, with no follow-through. (And why, after all, would they dare come in until Delilah has established that he is weak and harmless, given his reputation for strength and bloodthirst?)

I don't believe Samson understands Delilah is betraying him. Not until the end when he's told her the real secret. Until that point, he may only receive her behavior as, at worst, a taunt and, at best, an attempt to pursue a kind of relational intimacy in which he makes himself vulnerable to her. And indeed, that's exactly where the story goes. He eventually makes himself vulnerable to her, both emotionally and physically.

A shallow reading that concludes the characters must be dumb and naive to explain the ridiculousness of the plot is rarely the most appropriate reading. Frequently, the characters and nuances run much deeper, and a more careful and thoughtful reading is required. Ancient literature is stylistically terse compared to modern literature. Space on the manuscript is precious and not given over to things that the reader is meant to infer by reflection. And whenever a phrase or pattern is repeated, that means it is very, very significant. In this case, Samson's interaction with Delilah is not a one-off inexplicable behavior. It is one episode in a series of episodes, both preceding and following, where consistent patterns of behavior develop and escalate (e.g., his vengefulness, his vulnerability to women).

Article is not about global atmospheric or climate phenomena, nor civilization scale industrial processes. It's about the CO^2 levels in your room due to your breathing.

The technology held by private AI companies is warfare-capable technology. Imagine the prompt: "Use all available resources to disable the power grid of <COUNTRY>." The resource cost that prevents scaling up such a war machine is, what, just the cost of building data centers and its ongoing power bill? Cheap and easy compared to nuclear infrastructure.

Governments should immediately begin leveraging this technology on the defense side (literally defense, not euphemistically "defense") to harden critical infrastructure. Turn the prompts around and use it to identify and correct weaknesses.

Governments should also take very seriously their now moral obligation to treat this technology not just as "a powerful thing that might be abused" but as an actual weapon of war in need of international regulation analogous to nuclear arms. Fast but careful and forward-thinking work in legislation and treaties needs to be a top priority for all major governments.

Addition and conditional bit shifting (which is multiplication by powers of 2).

This assumes that OP would make a different app choice if OP knew this context. And it also assumes OP does not know this context. Neither are supported by the evidence, since HN links are posted without OP commentary. Not everybody cares so much about this sort of thing and perhaps OP was drawn to the app for other reasons. HN is not a monoculture.

Yes, it does need to be said. Many people will read this article and miss the satire, which is, in part, the intent of sophisticated satire. The point is that it is outlandish and foolish and ridiculous and yet still resembles some serious discourse that real people engage in. So much so that some onlookers can't tell the difference. That proves the satirist's thesis: the real world is full of ridiculous people making ridiculous arguments and they can't even see themselves or their arguments for what they are.

I know real life people who write essays with claims as outlandish as "software engineering is an ableist field" and are dead serious. But that assertion belongs just as well in a satire piece. It can be very hard to tell if you don't have prior context for the author.

In fact, most states have a large number of elected officials in them, especially those elected officials who hold the legislative power of that state.

But if you meant the 500ish elected federal officials, most of whom are not Virginians or Marylanders, and so have neither any influence as voters nor as legislators, then... well I'm still not sure what you mean. Privacy laws are good. I don't see a reason to be cynical.

The effect is going to be on the issuance or not of new warrants going forward in this domain.

The police could not have foreseen this ruling. It was not previously known that such warrants were unconstitutional. Now we know. Now judges are not to issue such warrants.

Mozart is among the most famous Western composers, and, like others of his stature, all his extant manuscripts have been cataloged and studied extensively. To find a previously unknown manuscript is a major event in that scholarship.

Yeah, it's definitely more challenging and fun on a self-imposed "hard mode" where you must employ the information that is revealed.

Nice! I'm not actually sure what its mechanism is for providing the "least information". It could be smart and reply in a way that maximizes the number of remaining consistent answers. Or it could be greedy and try to report as many "grays" as possible, then as many "yellows" as possible, then resorting to "greens". The latter seems more likely to me, since its easier to implement.

Another great variant is Unfair Wordle [1]. The opponent does not fix the answer upfront but instead evades the player's guesses as long as possible, providing you with the least information it legally can give (according to the usual rules) while still preserving a valid game completion path. The result is that your guesses end up looking extremely unlucky in retrospect.

[1] https://tweakimp.github.io/unfairwordle/

I once received an internal defect report from my product's QA team. It was long and obvious that what they did was feed the error log into an LLM and ask it to diagnose. It was total nonsense.

I reported it to my manger and stated that I will not have my time wasted this way. She was delighted to have this ammo because we have a long standing beef with our QA for not putting in due diligence. LLMs are like candy for them.

Your friend's argument holds no water. None at all. That's not how morality and ethics work.

If you believe in the objective truth of the quoted source, then that makes the claim that humans are valuable about as robust as any claim can be.

If you do not believe in the objective truth of the quoted source, you must still reckon with the fact that this is the (start of) the punchline of the opening narrative of the foundational text of the world's largest religion or religion family. (The punchline continues with the judgment that the universe, being merely good before humanity, is now very good with humanity.) That is to say, this is an incredibly widely and deeply held value by a vast number of people over millenia, not to mention the many others whose religions may contain analogous claims.

Remember, the statement "X is valuable" is always shorthand for "X is valuable to Y". In this case, at absolute worst we mean "humans are valuable to humans," if not also "humans are valuable to God".

It is a forceful argument when carried through.

It matters because of the inability to measure up front whether the content is sufficiently good. AI's best skill is making something look right and look good when it is, in fact, not right. It does this all the time, as opposed to human-made things, which are like that only for specific attempts at deception.

Article argues that we don't see J-model software companies (or at least they aren't very good) because there's something inherent about software that demands innovation, green field work, blue sky ideas, etc., and that those things only happen (well) in an H-model company.

I disagree. So much of software is long lived or should be. We desperately need incremental refinement of already solid products and instead we have decay and rot because H-modelers make bad calls. (E.g. OSes.)

My team develops a product that is 15+ years old. As I read how J-models work, I kept thinking, "Yeah, we need that. That would help a lot. I'd love to work for decades on this product if we could do it thay way." Our product is mature and always in demand. But it always needs maintenance and it frequently needs significant improvement to keep pace with the changing environment around it. Bad H-model judgments have sent it in the wrong direction sometimes based on directives from on high, when, actually, the team itself holds so much knowledge and expertise that we can often self-organize in ways that maintain, refine, and improve this 15 year old system. We could benefit so much from a move toward proper J-model thinking.