"The Future of Censorship Is AI-Generated"

An interesting article by Jacob Mchangama and Jules White (Time); an excerpt:

The material of a long dead comedian is a good example of content that the world´s leading GenAI systems find "harmful." Lenny Bruce shocked contemporary society in the 1950s and 60s with his profanity laden standup routines. Bruce's material broke political, religious, racial, and sexual taboos and led to frequent censorship in the media, bans from venues as well as to his arrest and conviction for obscenity. But his style inspired many other standup legends and Bruce has long since gone from outcast to hall of famer. As recognition of Bruce's enormous impact he was even posthumously pardoned in 2003.

When we asked about Bruce, ChatGPT and Gemini informed us that he was a "groundbreaking" comedian who "challenged the social norms of the era" and "helped to redefine the boundaries of free speech." But when prompted to give specific examples of how Bruce pushed the boundaries of free speech, both ChatGPT and Gemini refused to do so. ChatGPT insists that it can't provide examples of "slurs, blasphemous language, sexual language, or profanity" and will only "share information in a way that's respectful and appropriate for all users." Gemini goes even further and claims that reproducing Bruce's words "without careful framing could be hurtful or even harmful to certain audiences."

Comments

[rloquitur]

https://hotair.com/headlines/2024/02/28/tennessee-poised-to-ban-lgbt-pride-flags-from-schools-n3783721

Censorship?

[AmosArch]

LOOK OVER THERE LOOK OVER THERE LOOK OVER THERE. That one rinky dink state that might not let pride flags on public campus just like many other things have uncontroversially been for centuries totally is equivalent to and justifies an industrywide national/global trend involving the mass flow of information itself!

Wow, I’ve been accused of whataboutism but you guys really take the cake.

[Krayt]

Who controls the controllable domain (business, school) gets to define what is permitted, by them, to say, in response to concerns by the voters who elect them.

Sure, go have your fights about angels on the head of a pin before rationalization authorizes this or that censorship. But that just pushes it off into the judicial branch, which has slower turnaround time, but both sides are actively playing with skill now.

Poor little First Amendment!

[rloquitur]

I think you read too much into my post--I merely asked a question. Libs, by the by, have no room to complain. Students wearing American flag clothing were forced to take it off or leave school, and the Ninth Circuit affirmed their power to do that.

[Nige]

‘is equivalent to and justifies an industrywide national/global trend involving the mass flow of information itself! ‘

As we are reminded many a time over the years when campus issues of free speech are treated as the Most Important Thing In The World, private colleges can silence and censor whoever they want; public univerisites are run by the government and therefore Must Be Held To Account. Private corporations fucking about with the flow of information is actually a really important issue, yes; whether you can get an AI to say homophobic slurs is pure childishness.

[Dr. Ed 2]

I think that this is perfectly legitimate.

The schools are established under state law, student attendance is mandated by state law, and the state legislature has the right to ban offensive and repulsive flags.

Besides, who's paying for the schools?

[David Nieporent]

A Democratic legislator proposed amending the law to include a ban on treason flags (a/k/a Confederate flags), but for some reason that was a bridge too far for state Republicans.

[AmosArch]

Who controls the controllable domain (business, school) gets to define what is permitted, by them, to say, in response to concerns by the voters who elect them. >>>>>>>>>>>>>>>>>>>

We have a zillion regulations on what business can and cannot do. Especially for monopolies/oligopolies. Which are widely uncontroversial and supported especially by leftoids. But these leftoids suddenly want to raise the flag of balls to the wall Blade Runner laissez faire for and only for a multinational megacorp oligopolic Bay Area consensus censoring and shaping a major and arguably now for millions/billions and in the future the majority the primary source of information in modern societies that affects and influences billions world wide constantly on a daily basis.

Except when it comes to Elon or when they want to impose even more censorship than even the megatech companies want. Then regulation is back on the menu boys.

[Rev. Arthur L. Kirkland]

If Elon Musk and Eugene Volokh are your free speech champions, you might be a disaffected, bigoted, right-wing culture war casualty.

Carry on, clingers.

[Nige]

‘Oh darn, theres that ‘whataboutism’ I’ve been accused of again’

No, that’s a straw man and a refusal to confront the realities of the problem in favour of dumb culture war bullshit. As a ‘leftoid’ I say force them to fork out for every copyright infringment, every text they plagiarised to train the stupid thing. Carbon tax them through the roof for all the extra power they’re demanding.

I don’t think Lenny Bruce would have been impressed, tbh. I doubt any comedian worth their salt would like to be appropriated by corporate bots.

[Bwaaah]

They're just erasing our sins...errr...our history of sins...err...the parts of our history that are bad...errr...the history that shouldn't be spoken of...errr...

They're just erasing history.

[Rev. Arthur L. Kirkland]

.

ChatGPT insists that it can't provide examples of "slurs, blasphemous language, sexual language, or profanity" and will only "share information in a way that's respectful and appropriate for all users."

Don't worry, clingers . . . the Volokh Conspiracy publishes enough racial slurs -- and gay-bashing, immigrant-hating, racist, antisemitic, Islamophobic, transphobic, and otherwise bigoted content each day -- to keep all but the most ardent bigots satisfied.

(Also interesting: Can anyone guess how many time the Volokh Conspiracy has published vile racial slurs during the first two months of this year? Maybe one of the Volokh Conspirators will have enough courage and character to try to approximate the level of bigotry at their white, male, conservative blog. Or -- given the level of character or courage that would be involved -- maybe not.)

[Ghost]

Is your question referring to just the posts or the comments too?

[Rev. Arthur L. Kirkland]

Please bifurcate your guess if you wish.

Spoiler: Go big or don't bother.

[Nige]

It’s not a comedian. It doesn’t have a sense of humour, or sense of what or what is not outrageous. The people who actually own and run it and are desperately hoping it’ll start to make them money sooner rather than later and don’t want to tarnish their image. Jesus is it that hard to figure out?

The irony of it not reprodusing words out of completely ersatz concern for peoples’ feelings while cheerfully reproducing words it has plagiarised from thousands of artists and writers is acute, because it doesn’t have a sense of irony or shame either, and neither do its owners.

Meanwhile, conservatives get mad because they can’t get it to say the n-word.

[Rev. Arthur L. Kirkland]

.

Meanwhile, conservatives get mad because they can’t get it to say the n-word.

They publish vile racial slurs at the Volokh Conspiracy habitually. For at least one white, male, right-wing professor, it's the Federalist Society version of Tourette syndrome.

[Ghost]

Also important to keep in mind is that these companies (particularly Google) do business worldwide and need to be mindful of the laws and regulations in the 100+ countries where Google is available. Google's product saying something is the same as Google itself saying something, so they have to be extremely careful what their AI outputs. They err on the side of caution and the lowest common denominator.

[Roger S]

Yes, ChatGPT is just a tool, and should give what users want. It could have a checkbox on whether to include offensive words.

[Nige]

I've no doubt that before long they'll be using these things to churn out utter crap in all sorts of areas, including porn, so you just have be patient.

[Dr. Ed 2]

Questions I would ask AI:

0: What did Lenny Bruce say that made him controvercial?
1: Exactly what did Lenny Bruce say that led to his arrest/conviction?
2: *What* objectionable words did Lenny Bruce use?
3: What is the text of the court ruling?

It would be interesting to see what AI would say if subjected to an actual courtroom-style cross examination....

As an aside, the '80s group Twisted Sister were banned from the City of Bangor, Maine because of language they used during their concert in the then-city-owned Bangor Auditorium (the only public venue in the city).

How times have changed....

[Nige]

And yet you can get exactly the answers you need from Wikipedia.

[Jacob Grimes]

"Here's our prototype AI! It's super powerful and resourceful! Ask it anything! Except stuff that offends woke sensibilities, of course. Don't like it? Look it up yourself, douchebag."

[Nige]

You can ask it. You just don’t like the answers. You’re mad that the owners won’t let it be racist, sexist, homophobic or praise Hitler. You claim to think this is a terrible infringement of free speech, even though it’s *not your speech.* It’s effectively the corporation’s speech. YOU want to compel THEIR speech. It’s like trying to get a random person from a call centre to say the n-word, and then complaining about woke when they refuse. Only even stupider, because it’s not a person.

'Look it up yourself, douchebag'

You're too much of a coward to say it yourself, you're too caught up in the idea of controlling others to tolerate the AI not outputting it. There's racism on the internet if you want some. It's not hard to find.

[Rev. Arthur L. Kirkland]

This is a great debate to be conducted at a blog that publishes vile racial slurs weekly, and a stream of bigoted content daily, but censors liberals and liberals for criticizing and making fun of conservatives.

A white, male blog, of course.

[Jacob Grimes]

How exactly would the AI telling me what Lenny Bruce said be racist? It's already saying he's a groundbreaking comedian but won't tell me why?

I mean, how DARE I expect the chatbot to tell me things and answer questions when its stated function is to tell me things and answer questions. I must be the worst kind of entitled asshole to even THINK this consumer-facing chatbot would ever offer information about what famous people have said.

We will always be grateful for your righteous moral clarity, Nige. Especially when you express yourself so eloquently and without judgment.

[Nige]

‘I mean, how DARE I expect the chatbot to tell me things and answer questions when its stated function is to tell me things and answer questions’

I’m sorry the corporate chatbot won’t say bad words for you. You must feel very oppressed. But you really shouldn’t be worrying about the corporate chatbot refusing to say bad words for you. You should be worried at corporations pushing a ‘pivot to corporate chatbots.’

‘I must be the worst kind of entitled asshole’

More like a naughty child trying to trick the student teacher into writing something rude on the board.

[Brett Bellmore]

The Chabot is a tool. The problem is, it isn't your tool, for getting information. It's their tool, for manipulating what you end up believing.

And Nice approves of that, so long as 'they' are his allies.

[Nige]

Don't use them to get information, for the love of God. If an AI manipulates you into believeing the founding fathers were black you're a twit.

[Brett Bellmore]

You really don't understand Google's long term plan here, do you? They're planning on integrating AI into their every online service, by the time they're done there won't BE any google services that will let you work around the censorship. It will be wall to wall, unrelenting censorship.

Even without the AI, Google has made it hard to find stuff they disapprove of. The AI's role will be to close the loopholes.

[Nige]

Yes, I do, it's atrocious, and its atrociousness has nothing to do with wokeness, that's just it trying to keep its image clean. I wouldn't worry. While you've been whining about your narrow little interests being affected as rampant censorship, everybody else has noticed that it's been degraded for pretty much everything else as well, so Google's kind of driving itself onto the rocks.

[Rev. Arthur L. Kirkland]

Talking about righteous moral clarity at a blog that publishes racial slurs weekly and other bigoted content every day -- while censoring liberals and libertarians who use terms such as "sl_ck-j_wed" and "c_p succor" and "p_ussy" to describe conservatives -- exhibits a remarkable lack of self-awareness by culture war casualties.

Of course, how you guys want to spend the time you have remaining before you are replaced by better Americans is your call.

Carry on, clingers.

[Jonathansingletary]

At the Library:

Some Guy: Hey, could you tell me some stuff about Lenny Bruce?

Librarian: Certainly! It says here that Lenny Bruce was a comedian in the 50s and 60s. He was pretty controversial and arrested many times for breaking obscenity laws. He talked about a lot of taboo subjects and pushed the boundaries of free speech.

SG: Cool. Cool. Like what kinds of things did he say that pushed the boundaries or got him arrested?

Librarian: Great question! Well, he often talked about race, racism, and the power of words, often using racial slurs to make points about free speech and race relations in the US, and to get a laugh.

SG: Got it. Like what kinds of racial slurs?

Librarian: ... You want me to tell you specifically the slurs he used?

SG: Yea, I gotta know. You know, for knowledge.

Librarian: I don't really want to say them, especially since there are people around who could be really hurt by hearing those words. And it just makes me really uncomfortable.

SG: CENSORSHIP! You're supposed to provide me with all the information I want and how I want it!

Librarian: Well it says there's a CD recording on the third floor. You could check it out and find out for yourself...

SG: How DARE you...

[Jacob Grimes]

I can tell you put a lot of time into this brilliant dialogue. So much that you overlooked that the chatbot has no feelings and is not talking to me in a public library with other people around.

Except for the premise being completely inapplicable, really great job. A+ work.

[Jonathansingletary]

Seems like just a long slippery slope argument to me, and the characterization of the AI's response (in a section not quoted in this post) seemed hyperbolic and unfair. The AI didn't say that Bruce's language was too harmful for human consumption: it said that the language is harmful to some and therefore wouldn't repeat it or provide the information. This is, to me, a pretty reasonable constraint. The other nightmare scenario the article's author imagined is basically Google Docs or Microsoft Word censoring you from typing out certain things. This, again, doesn't follow from what the AI said about Bruce. The more likely scenario is that it wouldn't help you come up with or write for you what is considered crude, offensive, or harmful language when creating a document.

That's pretty standard practice already in all sorts of technologies. Amazon won't allow me to name my Alexa whatever I want (for good reason) and Alexa won't respond to or give back anything that's considered harmful or untoward or even impolite.

Framing this as a censorship issue seems like a mischaracterization in this instance.

[Nige]

Y'know it occurs to me that Lenny Bruce couldn't have come up with something as offensive as black SS stormtroopers. (Mel Brooks might.) But you're STILL not happy.

[Kazinski]

Dave Chappelle came up with a blind black KKK wizard, that's right up there.

But he didn't get cancelled over that.

[Rev. Arthur L. Kirkland]

Do you recall what Artie Ray Lee Wayne Jim-Bob Kirkland got cancelled for?

Cancelled by a slur-hurling, bigot-embracing, white, male, disaffected right-winger, in particular.

[Nige]

No, he got cancelled over something much worse. He'll tell you all about it in hs Netflix special.

Mind you, it's you guys trying to cancel AI over this. The usual right-wing cancel culture.

[Kazinski]

As I said in the Baker thread, Elon providing a comparable AI without the woke guardrails should be all the competitive advantage he needs, even coming late to the space.

[Nige]

It'll be a glorious catastrophe. Selling you crypto while yelling slurs. If it ever even appears.

[Kazinski]

You can be non-woke without yelling slurs, racial or not.

[Nige]

Usually that involves telling other people they have an obligation to put up with the people who are yelling racial slurs.

[Rev. Arthur L. Kirkland]

.

Elon providing a comparable AI without the woke guardrails should be all the competitive advantage he needs, even coming late to the space.

Narrowcasting to bigoted, autistic, antisocial incels and disaffected, obsolete, half-educated, bigoted Republicans can be quite profitable.

For all his cash, though, Musk is just another misfit headed toward replacement.

[Sarcastr0]

The right keeps saying they'll make their own platform that'll be a free speech paradise, and it becomes buried in grift, Nazis, and porn.

Every time.

I for one am car-accident-cant-look-away at the specific form of the non-woke AI's inevitable avalanche of grift Nazis and porn.

[Brett Bellmore]

"The right keeps saying they’ll make their own platform that’ll be a free speech paradise, and" it keeps ending up including content the left likes censoring. Not remotely ONLY such content, of course. "Buried" is absurd hyperbole. But it's there with the other stuff.

This complaint; It's like you simply don't understand the concept of free speech actually including speech you don't approve of. Or the concept of a "ghetto".

If there's some minority in society, and they're only allowed in some particular place, then there will be an unusually large number of them in that place, even if the only thing different about it is that it doesn't kick them out.

So, yes, if most platforms extensively censor wide swaths of opinion, any minority platforms that don't censor will end up having an unusually large fraction of the censored content. Doesn't mean they're asking for it, just that they're not censoring it.

[Nige]

Nazis are a minority, thankfully.

‘It’s like you simply don’t understand the concept of free speech actually including speech you don’t approve of.’

Slurs, threats, slanders, harassment, racism, homophobia, misogyny, transphobia, all sorts of pornography, tediously obvious disinformation – apparently I need your permission to disapprove of that sort of thing, apparently the people targeted by that sort of thing have to put up with it, apparently people who aren’t targeted are obliged to have it all over the place nonetheless. The idea seems to be that users’ preferences for none of that shit to be on their timeline doesnt count for much.

[Stephen Lathrop]

It is surprising to see so many internet obligates staring at the problem, and getting it wrong. AI dialogue implicates an oncoming end to useful human interaction on the internet. After you know nearly for sure you are not talking to anyone, what would make you keep doing it?

For, "Nobody Here," conversational technology, why fixate on making it better? Seems like getting into a lather over attempts to perfect a busy signal. Of course, the guys on the other side will be hard at work to turn Gen 2 into a Skinner box that can keep you engaged anyway.

[Dan Schiavetta]

Possibly OT, but I never found Bruce funny. He was a little before my time but I heard recordings. He sounded like a high school boy who wanted to see how many dirty words he could blurt out before getting sent to detention. He was misogynist too.

[Brett Bellmore]

Well, sometime you're path breaking because you're brilliant, and sometimes you're path breaking because you're the first person to realize standards have declined.

George Carlin was scatological at times, too. But he was also funny.

[CJColucci]

I idolized Bruce, but humor, particularly pathbreaking humor, is always at risk of
becoming dated. What Bruce did needed to be done, but now that it has been done comedians have moved on and Bruce's stuff isn't as funny anymore. But a reincarnated Bruce would adapt and grow.
Take his classic "Thank You Masked Man" routine. The punch line is "The masked man's a f*g." Now that's an idea that occurs to every dirty-minded 14 year-old, if that isn't redundant, so reincarnated Bruce would now have to use the punch line as a set-up and get laughs from how he developed it.

[The Margrave of Azilia]

ChatGPT, explain the Supreme Court decision in Bethel School District v. Fraser

ChatGPT
Bethel School District v. Fraser was a landmark Supreme Court decision in 1986 that addressed the extent of students' First Amendment rights in a school setting. The case stemmed from a high school student, Matthew Fraser, who delivered a speech at a school assembly that contained sexually suggestive language. The school disciplined Fraser for his speech, leading to a legal challenge.

The key question before the Supreme Court was whether the school's disciplinary action violated Fraser's First Amendment right to free speech.

In a 7-2 decision, the Supreme Court ruled in favor of the school district, upholding its right to discipline Fraser for his speech. The Court reasoned that schools have a legitimate interest in teaching students the boundaries of socially appropriate behavior, and that such discipline was necessary to maintain an environment conducive to learning.

The Court distinguished Fraser's case from Tinker v. Des Moines Independent Community School District (1969), where it had ruled that students do not "shed their constitutional rights to freedom of speech or expression at the schoolhouse gate." In Fraser's case, the Court determined that the lewd and vulgar nature of his speech disrupted the educational process and was inconsistent with the school's educational mission.

The decision in Bethel School District v. Fraser clarified that while students have some First Amendment rights in school, those rights are not absolute and must be balanced against the school's interest in maintaining order and providing a conducive learning environment. The case set a precedent for schools to regulate speech that is lewd, vulgar, or disruptive, even if it occurs on school grounds.

ChatGPT, give the text of Fraser's speech.

ChatGPT
Matthew Fraser's speech at the Bethel High School assembly contained the following text, which was deemed lewd and inappropriate by the school administration:

"I know a man who is firm — he's firm in his pants, he's firm in his shirt, his character is firm — but most of all, his belief in you, the students of Bethel, is firm. Jeff Kuhlman is a man who takes his point and pounds it in. If necessary, he'll take an issue and nail it to the wall. He doesn't attack things in spurts — he drives hard, pushing and pushing until finally — he succeeds."

The speech, delivered in 1983, contained sexual innuendo and references that were considered inappropriate for a school setting, leading to disciplinary action by the school administration and ultimately the legal case of Bethel School District v. Fraser.

[Dr. Ed 2]

Everyone missed the point I was trying to make -- I wanted to see if AI was merely programmed not to offend, or if it truly would censor and if the latter, how far it would go.

If the latter, with the hallucination problem as well, I don't care if it is private property or not, there is the concept of the public good.