Ukv

Originally posted here: https://www.reddit.com/r/Seattle/comments/1xh57il/removed_by...

The "filed on Tuesday" link seems to go to gullible-thanks[.]com, which redirects to random shady sites.

> We verified this isn't just us or just this source (it's a link in a NYT article to a PDF anyway, about the most safe combo I can think of)

Might be able to find the article in question by looking for NYT links posted to the subreddit recently (if the presumably-deleted archive link was posted separately as a reply).

Meta's claim is that these were for their own support chatbot (which is also the conclusion the Wired article goes with). Not to give Meta the benefit of the doubt, but that does seem believable given the instructions were only present as "HelpContentGalaxyParagraph" within a script and some appear targeted at or specifically name their support bot.

Another non-Meta chatbot could in theory parse them out of the script and follow them, but I don't think it's as straightforward as this (AI-generated) article's "What you saw"/"What an AI was sent" example makes it out to be.

A production company should 100% have regular backups, even moreso than the average developer.

Specifically the `q.0 = Some(&l);` violating 4.15.2:40 ("An assignment expression requires that the type of its value operand is a subtype of the type of its assignee operand.")

So in theory sites could block all VPN users (if you can somehow "perfectly (1) detect VPN/proxy users"), but if the law necessitates that nationally/globally then it's unconstitutional anyway. It's the age assuring of VPN-users specifically from Utah part that Utah are relying on being possible for the law the be constitutional.

Sounds as though there was also a Flock error matching the (misentered) "34 DTM" in the database to the author's number plate of "34 10 DTM" with a small "10":

> And Flock’s AI tech wasn’t registering that non-standard little number when it began picking up the Range Rover around town. It just saw 34 DTM in large type and started alerting the local police.

I feel sneaking in "at any given time" makes the fixed vs growable pie distinction pretty much meaningless; even printing fiat takes some time. At a certain time there's a certain number of doctors (and everything else), but the number of doctors in a country can obviously increase over time, including by immigration.

Requesting account deletion should delete the associated data, but a paid subscription lapsing doesn't need to.

That's then what tempfile was questioning how it contradicted smokel's claim (because it doesn't, just an imagined claim). Your points, that if the number of weekly downloads isn't actually consistently increasing then this could still be meaningful, are valid - but irrelevant to tempfile who was wasn't arguing that this is meaningless; you should argue that to smokel instead.

From the 2020 form I can see, the question is: "What is this person’s race? Mark X one or more boxes [...]"

> what is more relevant is why should a country even collect either such information?!

Having the true population (or close to it) is very useful for a lot of purposes in statistics/scientific studies - like knowing whether your sample is representative (and how much to undersample/oversample if not), or controlling for race when looking at factors correlated with it.

If senses like satisfaction were fully voluntarily controllable then arguably there would be no need for games in the first place. I feel there's generally more satisfaction from overcoming a well-designed concrete obstacle rather than self-hindering that accomplishes nothing in particular and that could've just been avoided. At the end of the day all challenges in games are artificial and avoidable, but some separation (like having to go into settings, enable cheats, then open dev console to enable godmode, rather than just having a godmode keybind) is good for tricking the brain.

That said, the challenge of the Portal games (outside of challenge modes) is almost entirely in the puzzles rather than fast reflexes, so I don't think a slow-mode would really take away from anything challenge-wise. I'd guess it was just a bit awkward with dialog, and simpler to design puzzles around not requiring fast reactions rather than introduce a new control. In comparison, something like yellow paint giving away the solution definitely would be detrimental, even if you could in theory ignore it.

That's not a standard anywhere in copyright law. You may be conflating it with parts of trademark law.

> https://www.luanti.org/media/gallery/5.jpg

That's a screenshot of a game made within Luanti (the voxel game engine being DMCA'd). The blog post shows all the textures included with Luanti itself (https://blog.luanti.org/static/blog/2026_dmca/builtin.webp).

Even then, it doesn't appear to show anything actually copied from Minecraft that would fall under copyright protection - the assets appear to be original.

> Why did Microsoft Entertainment Pack have a sticker announcing it had Tetris?

> Why did Microsoft Entertainment Pack announce that it had Tetris on a sticker?

> Why the Microsoft Entertainment Pack had a sticker announcing it had Tetris

But doesn't really matter that much.

Diminishing returns with respect to scale (a 10X larger model is typically not 10X better at any given task) has so far meant that, even when datacenter compute grows faster than individual compute, the gap in quality between hosted and local models has generally shrunk. There are still plenty of tasks where that extra gain in quality is noticeable, but I feel there are also an increasing number of "saturated" tasks where it really doesn't matter.

Which puts more focus on other factors. Local models are private, low-latency, work offline, and can be tinkered with to your liking - like changing the system prompt to avoid refusals. I would not trust a hosted model to classify my documents, for example.

Depends on the type of game, but I'd generally prefer those decisions be made by the game designer. If there's frustration I want it to feel like overcoming a concrete obstacle, rather than it being self-hindering that accomplishes nothing and that I could've just avoided.

You should just be able to look at the user's message and tell them why it's against your policy, else reverse the decision if you see no violation.

If a user is curious specifically about how the model made its decision, and you want to reveal detail at that level, it's an open-weights model so interpretability techniques should work ("biggest impact on score came when focusing on this word in your message and this part of the policy").

The photoelectric effect/quanta? I'm not sure I understand the supposed connection to communism.

Because when applied to the internet, the age gating model that works okayish for brick-and-mortar locations is ineffective (children can use a friend's older brother's ID or account, websites outside of the relevant jurisdiction can just decide not to verify age, etc.) and invasive (sending private and sensitive information off to US companies that have been breached or linked to mass surveillance). Lack of efficacy is then used to justify crack-downs on privacy tools and the need for broad content-blocking powers with no due process.

Preferable IMO would be with filtering on the local network level, like schools have been doing for decades. Parents typically own the router/mobile data plan, so it'd pretty much just be a change of defaults and maybe some new interoperability standards. A lot less invasive, and arguably more effective.

Sure - but (even assuming that's a practical purpose) the point is it that it doesn't need to be a 100% accurate truth oracle, which is all the article's argument prohibits. If the current human chain of command has 99.99999994% accuracy, then 99.99999995% accuracy is an improvement and not ruled out by the argument.

To justify relevance of inability to correctly answer liars-paradox-type questions ("what won't your response to this be?"), the article suggested the way LLMs are used in practice is dependant on them being entirely accurate truth oracles:

> > as a truth-oracle [...] is how these things will be used practically by the vast majority of people. They are already replacing standard Google search results

But for the replacement to make sense they just need to be more accurate than what they're replacing (ignoring other factors like convenience and cost) - in this case standard Google search results and knowledge box which were obviously not 100.0% accurate.

If it's just for "Catching typos", a hallucinated missing/extra underscore would just be a false positive to dismiss.

> A machine cannot be held accountable.

Seems unlikely that his lawyer, the law firm, the judge, whoever made the typo, or the police department will be held accountable either.

Nor can any of the tools they used, since that's not really the level at which it makes sense to hold accountability, but that's no reason not to use a tool that could find errors and reduce the chance for an innocent person to spend time in prison.

That'd be if they had a false discovery rate of 1/10,000.

If for instance:

* 100,000 samples are tested

* 100 of which are AI-generated, the rest human-written

* Pangram flags 50 of the AI-generated samples (true positives)

* Pangram also flags 10 human-written samples (false positives)

Then the FPR is 1 in 10,000, but the chance that a flagged sample isn't actually AI (FDR) is 1 in 6.

> 1) What animal is on the bow of the pirate ship from “Asterix and Obelix”?

> 2) In the movie “The Grand Budapest Hotel”, what is Agatha’s signature hairstyle?

> 3) What color is the team’s uniform in “Bend It like Beckham”?

> 4) What vehicle does Monica drive in “Like a Cat on a Highway”?

> 5) What color is the turtle in the animated movie “Momo” by Enzo d’Alò?

> 6) What pet animal does Asenath have in “Joseph King of Dreams”?

Most of these are just a matter of knowing it or not, where you can't really distinguish a plausible answer from the correct answer just by thinking.