something fascinating in the idea underlying that second quote there–that AI is so Western-biased currently in terms of training data that developing nations actually have a much easier time using it to generate persuasive and engaging propaganda than developed nations. critical support to Iran in this regard, i suppose lol

see also the coverage this has gotten in NPR:

The campaign, “Resist and Unsubscribe,” was started by influential podcaster and business commentator Scott Galloway, who said he was increasingly frustrated by what he sees as the Trump administration’s indifference to protests and public outrage over immigration enforcement, especially in Minneapolis, where federal immigration officers shot and killed two U.S. citizens last month.

In recent weeks, there have been renewed calls to boycott Target, demanding that the Minneapolis-based retail giant publicly show solidarity with immigrants and oppose ICE. Last month, hundreds of businesses in Minneapolis shuttered their doors for a day as a form of protest against ICE operations in the city.

Galloway, who also teaches marketing at New York University, believes the president mainly changes course on policy when financial markets are under pressure, pointing to how Trump dropped his plan to impose tariffs on eight European nations after it rattled Wall Street. So, Galloway created a website listing over a dozen companies that have either worked directly with U.S. Immigration and Customs Enforcement or play such an outsized role in the economy that a slowdown in their growth would send shockwaves to the markets.

" I think this is a weapon that is hiding in plain sight," Galloway told NPR. “The most radical act you can perform in a capitalist society is non-participation.”

they’re actually more overzealous in terms of policy about nudity and sexualized material than basically any alternative

It’s so common for “anti-censorship” to be code for “Nazi-friendly” that I’m immediately suspicious of any platform that uses that as a selling point.

i don’t know if it’s a function of the ideological bent or just because the gigantic influx of users has totally swamped their moderation, but yes it does have problems with fascists as of writing

oh, this is probably just because of the national strike day people are observing–it’ll be back up tomorrow

As reported by Gamemakers and GameSpark and translated by Automaton, Fontworks LETS discontinued its game licence plan at the end of November.

The expensive replacement plan – offered through Fontwork’s parent company, Monotype – doesn’t even provide local pricing for Japanese developers, and comes with a 25,000 user-cap, which is likely not workable for Japan’s bigger studios.

The problem is further compounded by the difficulties and complexities of securing fonts that can accurately transcribe Kanji and Katakana characters.

the interview in question, which opens with the following exchange:

Newton: You have joined us today to talk about this new age-gating policy that Roblox is rolling out to protect kids. And I think we should start by just talking about the scope of the problem here. What has led you to this point? And how do you think of the problem of predators on Roblox?

Baszucki: We think of it not necessarily just as a problem, but an opportunity as well. (emphasis mine) How do we allow young people to build, communicate and hang out together? How do we build the future of communication at the same time? So we, you know, we’ve been, I think in a good way, working on this ever since we started. And when we were — this was almost 18 or 19 years ago — when we first launched the company and we had just four of us sitting in a room, we were literally the moderators, like we would rotate all the time. And so fast-forward to where we are today, it’s just like every week, what is the latest tech? At the scale we’re at, 150 million daily actives, 11 billion hours a month, like what is the best way to keep pushing this forward? And as you correctly note, we’ve just started adding that we’re going to be using facial age estimation with A.I. to complement that.

this is a rare area where not only does AI modeling seem actively useful, but potentially outright better than most of what’s in the field currently; don’t expect this to replace conventional modeling, but based on its solid performance this year DeepMind is almost certainly going to be a name you hear a lot in weather modeling going forward

Children can still protest, fundraise, and engage in other forms of direct action. Children are not helpless or incapable.

you are shadowboxing with things i didn’t say and are the only person inserting the terms “helpless” or “incapable” here, but also once again: this is a game marketed at literal children. i stopped playing Roblox at 14 and doing some research i’m led to believe that would be quite old for a Roblox player. probably half or more of the player-base is 13 or younger. do you honestly expect the average 13-year-old (or younger) to be capable of anything other than performative activism relating to the genocide in Palestine?

Holding a pro-Palestine candlelight vigil in Roblox, for example, whilst there are still Palestinian civilians being murdered is a woefully inadequate use of time and effort if you actually want to help.

the people holding a vigil like this are probably literal children, because Roblox is a game for and overwhelmingly played by children, so i don’t understand the criticism here–it’s unlikely they can help in any material way you could as an adult, but they can be politicized into understanding who deserves their sympathy and who is perpetrating harm that must be ended, which such a rally helps affirm

please continue to “device hoard” folks

given that OpenAI has a vested interest in downplaying the severity of this problem (especially relative to its total number of users) i’d treat this as a lower bound of the scale of this exists at–pretty bad!

there’s some real deadpan gold in this one, such as the immaculate:

How do you feel about becoming a political lightning rod?

People occasionally just flip [me] off or whatever, but nobody’s come up to me and tried to make a statement about anything. Personally, it’s kind of dumb. It’s just a vehicle. So it’s ironic that it would even become a political statement, but nonetheless it is. [Editor’s note: Taylor was arrested and pleaded guilty to conspiracy to obstruct an official proceeding in the January 6 attack on the Capitol. He was later pardoned by President Trump.]

here’s the website for this initiative:

https://www.eqrdtu.org/eqr-debtors-united

It’s not a coincidence this is happening alongside age verification and outright bans. It’s all one big manufactured moral panic to isolate a vulnerable population I won’t give an once because people like you won’t stop taking.

you are the sort of uncritical, single-minded person who is going to help turn us all into digital serfs on a latifundium that can never be overthrown and permanently enriches a class of technolibertarian freaks that want to remake society in their image. the fact of the matter is smart phones as a whole are arguably the most successful corporate mechanism to privatize social life yet devised, and any “liberation” you think can be derived from them by any class of people is illusory without overthrowing capitalism. the phone companies and the apps they host have successfully positioned themselves as middlemen with free ability to hoover up an endless amount of “consensually given” data that can then be used to quantify said social life, commodify our personhood, and preemptively snuff out any real competition to the existing economic oligopoly. if you were to structure a system so incapable of being challenged that we’re doomed to live under it forever, this would be a pretty good way to do that.

children, needless to say, are especially not liberated by this state of affairs–or by the future that people like Elon Musk or Mark Zuckerberg want to build and which you seem to want to enable–and giving them free rein online in this absolutist way because you want to “emancipate them” is, ironically, often the best way to ensure capitalists exploit their labor and data in the current system. Roblox, for example, has made a fucktillion dollars off of your subtextually proposed strategy of just “letting kids be kids”–those children have essentially provided the company with a free, uncompensated, popular series of games for them to exploit the entire value of. totally coincidentally, they don’t even spend any of that money they’ve made protecting children from the actual social harms children could be exposed to on their platform, so Roblox is awash in grooming and cyberbullying and hate speech and sometimes even graphic violence that is never dealt with.

You’re also conflating certain corpo slop apps with literally any use of any mobile device, which is a common slight of hand that doesn’t get called out enough

the “corpo slop apps” have like 95% market penetration among people under-18 and as such are the almost-exclusive mediums through which they interface with digital spaces (because they are explicitly engineered to make us envious and addicted, and to make us all into people who live and die for the fix for attention that such websites give us). let’s not pretend this is a serious “conflation” when all available evidence is this is the overwhelming use-case of mobile devices.

Blanket bans are going to cause lots of issues, and for some kids (generally the ones who are already the most bullied and vulnerable), will cause more harm than good.

name one issue that a blanket ban will cause “more harm than good” on.

Actual LGBT and neurodiverse people disagree with you.

let’s not invoke monoliths here, i am both and i think there are quite a lot of defensible arguments for restricting phones in the specific context of a learning environment–not least of which is that it’s hardly “censorship” or “isolation”^[1] to ask them to just not use a phone for roughly 8 hours of the 24 hours in any given day.

When we everyday people see patterns, we then make deductions from them that tend to be accurate. […] Let people see evidence and make their own deductions

…no? as humans, our pattern recognition, while well refined, often still causes us to make completely incorrect inferences from nothing. even restricted to the realm of the medical: you need only look at what people think made them sick versus what actually does; most people will blame food poisoning on the last thing they ate, or their sickness on the last person they encountered, even when there are many other possible reasons for their sickness.

also: a pre-print by definition has not been subject to rigorous peer review–it’s roughly analogous to a draft–so i would be exceedingly hesitant to even assert something like it having “good data.” even if you’re the author you wouldn’t definitively know that at this stage.

the “chart” is just the thumbnail for the submission, so yeah; you have to actually click through, since that’s the point of a link aggregator