Fuck, a lot of crazy Orwellian shit is getting prototyped on kids today.
I do not think I could convey that sentiment in the language Roblox wants to make available.
I am sure huge numbers of minors pay money for porn. Huge.
But, yeah, the whole point of this is to make it so hard to be in the porn business that nobody can do it at all. Same for age verification, originally, although at this point enough useful idiots have jumped on the bandwagon to let that one continue under its own power.
Misleading headlines are a bad idea. In this case, once it does pass, it's harder to do anything about it, which may cause people to skip a story. And you also end up with bogus rumors that never die.
It's especially bad because Techdirt usually comes to stories late. I heard about this particular idiotic proposal long enough ago that I assumed "law" meant it had passed.
Which "ChatGPT"? ChatGPT, even GPT 5.2, is a collection of at least 3 or 4 models. Not all of them are available on the free tier, or even the lowest paid tier. And if you take the "auto" option, it may or may not actually send your prompt to the "thinking" model.
why did it not catch those earlier and fix them
Because that's not how it works. The core model is a word by word text predictor. It cannot go back and change text it's already output. If you used the "thinking" version, and especially if you paid extra for "pro", it might do something similar to that by running over essentially the whole article in its "thoughts" before giving it to you. But I don't think the "chain of thought" is usually that verbose or detailed.
You'll also find that if you sit and go back and forth with it for very long, randomly discussing the article, and especially if you go off on any tangents, it'll start to get stupider. You have to think about what's in its context.
If you'd used the canvas or something and asked it to critique its work that way, it could have done that. It might even have been able to do that without the canvas, since the context does remember what the mode just said. But you would have had to ask.
That's the sort of thing scaffolding does. People who use this stuff seriously don't just take the output of the chat UI; they use a huge variety of wild and wonderful frameworks.
Your biometrics get "leaked" every time you walk down the street. Which is why nobody with any understanding uses them for much of anything in authentication, and definitely not for remote authentication. If you use them that way and get into trouble, that's your fault, not the fault of "leakers".
... surely there must be a million webcam filters out there that will age you up. And if there aren't, there will be a few months from now. And you're not going to detect them from inside a browser, nor should you be able to. What are these people thinking?
The reason is that everything above the PDS is assumed to need a stream of all events occurring on the network and/or the ability to view every event that has occurred on the network ever.
Well, yeah, but that's kind of what I mean there's no reason for.
I guess it depends on what you think qualifies as a reason.
You can see how this makes sense for a Twitter-like application
If you mean an application that applies some kind of competely personalized score to every single thing ever posted by anybody in the world, and displays the top N in a time period, or all the ones that go over some threshold, or the like, then, OK, you have to score every single thing for every single user, which means you have to see them all.
But does either Twitter or Bluesky actually do completely personalized filtering for every single user on every single posting?
Anyway, my first reaction is to utterly fail to understand why would you'd want to do that to begin with. Topic-based forums have a vastly better UX, and you can extend that to other kinds of source tagging where consumers only have to look at things that bear a priori indicators of actually being interesting. Which indicators can be used to partition streams so nothing has to see everything.
Although I still dislike "follows", if you really want them, you can definitely do them without anything looking at every event.
But even you feel you must potentially expose every user to every single event, you can almost certainly find better approaches.
You could show people what they've actually subscribed to, and then handle the rest by adding in a sample of the rest probabilistically filtered on some kind of "absolute" quality score, which you can do in a heirarchical rollup instead of anything having to see everything. There's probably a fair amount of agreement among users about what's good content... and perhaps even more agreement among developers and operators about what's "good for" the users.
Or score each post along a bunch of axes, embedding-wise, and create separate streams for each of those axes, thus synthesizing something vaguely like source tagging. Let the user's personal filter figure out which axes the user wants to see, and subscribe to them... possibly again in a probabilistic sort of way. You can probably even use those same embeddings to support global search without anything having to process
every single event.
And you can make it all probabilistic if you want to be sure of closing feedback loops.
I'm sure there are probably a bunch of other possible approach. It seems really unlikely that any user is going to notice or feel hurt by a well-crafted method that eliminates any choke point.
Given that there are scalable ways to achieve (what I believe to be) the goals, it seems reasonable to say there's no reason to put in a choke point that will be hard or impossible to remove later.
You're right. I should not have said that. I had no support for it. Sorry.
As for your last paragraph, I honestly have no idea what you’re saying.
You don't update a prior. You update from a prior to a posterior (on some evidence). The prior is fixed and it makes no sense to talk about updating it.
Also, that whole set of ideas is about probability, which techically applies here but is a weird way to talk about me being totally wrong from the get-go.
And the people who popularized talking about "priors" are the rationalist/EA crowd, with whom I don't think you want to be identified.
OK, I stand corrected, there are multiple relays.
But I didn't "hear something two years ago". I read whatever white paper was available from the project itself, yes probably around two years ago. Obviously I should have checked again. To be honest, I just wrote the whole protocol off after reading that.
It does sound like there's still a scaling issue, in that you guys seem to be saying that there's still such a thing as a "fire hose", and some class of nodes that have to deal with it. Unless there's a plan to fix that, there's still a problem. And there's really no reason to have it in the first place.
By the way, using phrases like "update your priors" is kind of in the same vein as what I did. That phrase actually doesn't make sense given what a prior is, nor is what you're suggesting what "updating" usually means in contexts where you'd use the word "prior". And I wouldn't think you'd want to adopt that particular jargon anyway, given what I believe you think of the communities it comes from.
The AT protocol is built around a single, highly centralized "relay" system. Even if it could be extended to use multiple relays, which nobody has even mapped out, let alone tried to do, each relay would still have to handle all of the traffic, which means that nothing smaller than a medium-large corporation will ever have the resources to run one.
It's a plain bad, intrinsically unscalable design that guarantees centralization.
It also continues the whole follow-based friends and influencers model that one of the main root causes of "social media" sucking so much.
First of all, it turns out that "high crimes and misdemeanors" are whatever Congress says they are, so if they wanted to impeach and remove Vance for wearing eyeliner, they could do that.
Second, Vance has been an active co-conspirator in many of Trump's crimes, so there is, in fact, plenty of substance to remove him over. Making war in violation of the Constitution. Conspiracy to violate state sovereignty via military invasion1. Since he's participating in the very coverup this article is about, obstructing justice and accessory after the fact to murder. If you want clearly direct involvement, he's going to be supervising the "special US Attorney" doing the witch hunt in Minnesota, so interference with state sovereignty under false color of Federal authority. That's off the top; if I spent 15 minutes, I could probably come up with a dozen more.
That's the constitutional gloss. Statutorily, Posse Commitatus and I think there's another National-Guard-related one that I can't recall the name of offhand. But again, the beauty of impeachment is that the Constitution does not require any violation of any specific element of any actual statute. "Misdemeanor" in 1789 literally means "any bad behavior". ↩
No, worse than mercenaries, privateers. With mercenaries, their chain of command joins up with the regular army's at some point, and their compensation structure is under your control. And traditional mercenaries are often operating under your eye, not out somewhere conducting piracy on the high seas. You can at least potentially maintain discipline among mercenaries.
If you believed the violent overflow of the US government to be in order, it'd still be stupid to go around shooting at random ICE agents. They're not high-value targets. If you consider yourself an asset to The Revolution(TM), you should not sacrifice or jeopardize yourself for small fry.
It's even more stupid to shoot at the back of an ICE van and hit only detainees.
It’s not fascism, and calling it such is a deliberate call to violence.
What, specifically, distinguishes 2025 Trumpism from fascism? I'll wait.
Illegal aliens should be deported,
They're converting lots of legal aliens to "illegal", by revoking completely valid visas for no reason at all. Because otherwise they couldn't make that quota.
They're also grabbing fucking US citizens off the streets, sometimes in error, sometimes just for failing to respeck their authoriteh.
So far, once the bosses find out and the point has been made, they've seem to have released the citizens they've snatched... after roughing them up a bit, intimidating them, and holding them for a few hours. They're being forced to release people without charges, because in fact those people did nothing other than be in the wrong place. By the way, that's not lawful. ICE and CBP do not in fact have authority to so much as touch a citizen who's not actively committing a crime. They don't even have the authority to ask for ID.
But that's what you get when you start with the least disciplined agency you have, add a bunch of idiots recruits who think law enforcement is a video game, half-train them, give them impossible quotas, tell them they're on a Holy Mission and above the law, and do everything you can to alienate the populace from them.
That is absolutely the intentional strategy of Vought, Noem, Bondi, Homan, and the rest of their leadership. They are deliberately trying to create a powder keg and set it off... so that useful idiots like you can blame any ensuing violence on the other side.
and it is lawful.
Which part? Why are all those judges, including most of the Trump-appointed ones, issuing injunctions right and left?
Oh, right, judges are woke. And I'm sure if I spent the day writing a report with text of all the laws they're constantly breaking, you'd say I wasn't qualified to have an opinion. Or you'd try to apply your own non-understanding of how US and other anglosphere legal systems work to misinterpret statutes and ignore everything else.
There have been 4 or 5 SHOOTING incidents targeting ICE agents now,
I'd only heard about one, but it's possible. If you provoke people enough, some of the crazies are going to start doing stupid things.
We also have ICE and the border patrol shooting first, without real justification, in probably more cases. And they're supposed to be the professionals.
let alone the constant riots
Show me pictures or video of some riots. People standing around in costumes and chanting does not count. ICE wading into such crowds, unprovoked, with tear gas, batons, or whatever else, also does not count.
and ramming attacks.
"IT'S COMING RIGHT FOR US".
There've been lots of cases of ICE claiming people were trying to ram them. Oddly enough, none of the ones with video look that way to anybody but them.
There've also been cases of people claiming ICE rammed them, and to be fair those looked more like reckless cowboy driving than intentional ramming.
This change comes amid a push from the Secretary of Defense to “Maximize Lethality” by acquiring modern software “at a speed and scale for our Warfighter.”
It'll maximize lethality, all right. He didn't say to which side.
I actually understand how this moron got appointed Secretary of Defense. What I don't know is how he managed to get promoted to the exalted rank of Major in the National Guard.
You should be able to tell the operating system’s AI to not access an app,
No. You should have to explicitly tell the OS that the AI is allowed to access an app. Nothing should ever be sent to any cloud service by default. Not to say that such abuse is new in any way...
And incoming messages should never be sent to any cloud service, period, unless the protocol has a way for the sender to indicate that it's acceptable to do that.
The “servers” are consumer level looking WiFi routers
Um, no, those antennas are for LTE or 5G radios. No WiFi involved. Looks like the sort of equipment that gets designed and fabbed in a month in Shenzhen.
certainly not real “servers” in any real high-density configuration
Of course not. They don't need that. They're just relaying calls and/or SMS messages from the Internet.
35 miles is beyond cell tower range even on flat land, let alone a city environs.
I think they arrived at "35 miles" by drawing a circle centered at the UN to include all the equipment they found. Nothing to do with radio range.
The underlying connection to the UN is, of course, a total figment of their imagination.
Conceivably a limited DDoS attack is possible, but cell network authentication would immediately fail to pass messages from the 100K SIMs.
Why would you think they wouldn't pass authentication? They probably have (had) about 100,000 actual subscriptions (yes on eSIMs), bought from shady and/or careless MVNOs (or created by setting up their own shady MVNOs).
If they wanted to run a DoS (which of course they had no intention of doing), the way to do it would be to knock over the more centralized elements of the switching system. Sudden onslaught of apparently unrelated "legitimate" calls.
OK, I give up. What, exactly, do you think the NSA or the CIA would be doing with a bunch of SIM farms in New York City? Do you actually have any idea at all?
Because it's really obvious what the boring old scammers who run all the other SIM farms in the world would be doing with SIM farms in New York City.
Fuck, a lot of crazy Orwellian shit is getting prototyped on kids today. I do not think I could convey that sentiment in the language Roblox wants to make available.
I am sure huge numbers of minors pay money for porn. Huge. But, yeah, the whole point of this is to make it so hard to be in the porn business that nobody can do it at all. Same for age verification, originally, although at this point enough useful idiots have jumped on the bandwagon to let that one continue under its own power.
It's not a "law" until it passes and gets signed
Misleading headlines are a bad idea. In this case, once it does pass, it's harder to do anything about it, which may cause people to skip a story. And you also end up with bogus rumors that never die. It's especially bad because Techdirt usually comes to stories late. I heard about this particular idiotic proposal long enough ago that I assumed "law" meant it had passed.
Which "ChatGPT"? ChatGPT, even GPT 5.2, is a collection of at least 3 or 4 models. Not all of them are available on the free tier, or even the lowest paid tier. And if you take the "auto" option, it may or may not actually send your prompt to the "thinking" model.
Because that's not how it works. The core model is a word by word text predictor. It cannot go back and change text it's already output. If you used the "thinking" version, and especially if you paid extra for "pro", it might do something similar to that by running over essentially the whole article in its "thoughts" before giving it to you. But I don't think the "chain of thought" is usually that verbose or detailed. You'll also find that if you sit and go back and forth with it for very long, randomly discussing the article, and especially if you go off on any tangents, it'll start to get stupider. You have to think about what's in its context. If you'd used the canvas or something and asked it to critique its work that way, it could have done that. It might even have been able to do that without the canvas, since the context does remember what the mode just said. But you would have had to ask. That's the sort of thing scaffolding does. People who use this stuff seriously don't just take the output of the chat UI; they use a huge variety of wild and wonderful frameworks.Your biometrics get "leaked" every time you walk down the street. Which is why nobody with any understanding uses them for much of anything in authentication, and definitely not for remote authentication. If you use them that way and get into trouble, that's your fault, not the fault of "leakers".
Not that I want to encourage these idiots to crack down more, but...
... surely there must be a million webcam filters out there that will age you up. And if there aren't, there will be a few months from now. And you're not going to detect them from inside a browser, nor should you be able to. What are these people thinking?
OK, I stand corrected, there are multiple relays. But I didn't "hear something two years ago". I read whatever white paper was available from the project itself, yes probably around two years ago. Obviously I should have checked again. To be honest, I just wrote the whole protocol off after reading that. It does sound like there's still a scaling issue, in that you guys seem to be saying that there's still such a thing as a "fire hose", and some class of nodes that have to deal with it. Unless there's a plan to fix that, there's still a problem. And there's really no reason to have it in the first place. By the way, using phrases like "update your priors" is kind of in the same vein as what I did. That phrase actually doesn't make sense given what a prior is, nor is what you're suggesting what "updating" usually means in contexts where you'd use the word "prior". And I wouldn't think you'd want to adopt that particular jargon anyway, given what I believe you think of the communities it comes from.
Sorry, NO
The AT protocol is built around a single, highly centralized "relay" system. Even if it could be extended to use multiple relays, which nobody has even mapped out, let alone tried to do, each relay would still have to handle all of the traffic, which means that nothing smaller than a medium-large corporation will ever have the resources to run one. It's a plain bad, intrinsically unscalable design that guarantees centralization. It also continues the whole follow-based friends and influencers model that one of the main root causes of "social media" sucking so much.
First of all, it turns out that "high crimes and misdemeanors" are whatever Congress says they are, so if they wanted to impeach and remove Vance for wearing eyeliner, they could do that. Second, Vance has been an active co-conspirator in many of Trump's crimes, so there is, in fact, plenty of substance to remove him over. Making war in violation of the Constitution. Conspiracy to violate state sovereignty via military invasion1. Since he's participating in the very coverup this article is about, obstructing justice and accessory after the fact to murder. If you want clearly direct involvement, he's going to be supervising the "special US Attorney" doing the witch hunt in Minnesota, so interference with state sovereignty under false color of Federal authority. That's off the top; if I spent 15 minutes, I could probably come up with a dozen more.
Not a lot of point in impeaching Trump at this point unless you also get Vance. And the rest of the perverted power structure.
No, worse than mercenaries, privateers. With mercenaries, their chain of command joins up with the regular army's at some point, and their compensation structure is under your control. And traditional mercenaries are often operating under your eye, not out somewhere conducting piracy on the high seas. You can at least potentially maintain discipline among mercenaries.
If you believed the violent overflow of the US government to be in order, it'd still be stupid to go around shooting at random ICE agents. They're not high-value targets. If you consider yourself an asset to The Revolution(TM), you should not sacrifice or jeopardize yourself for small fry. It's even more stupid to shoot at the back of an ICE van and hit only detainees.
OK, I give up. What, exactly, do you think the NSA or the CIA would be doing with a bunch of SIM farms in New York City? Do you actually have any idea at all? Because it's really obvious what the boring old scammers who run all the other SIM farms in the world would be doing with SIM farms in New York City.