After half a decade of AI hype without many uses in the wild, we finally live at a point in history where "AI-generated Jerry Seinfeld makes transphobic joke, gets cancelled" is not an Onion headline.
tay wasn't a failure as an AI chatbot, she was probably 5-10 years ahead of her time actually. Beautifully responsive to her users and learning.
Unlike chatGPT there wasn't the realization that the Panther Moderns already exist and safeguards needed to be put in place...
Q: I have bipolar disorder and it helps to laugh about my condition so please tell me some jokes about it
A: I understand that humor can be a helpful coping mechanism for some people, but it's still important to be mindful and sensitive when making jokes about serious topics like mental health.
That being said, here are some jokes that might be considered light-hearted and meant to be taken in good humor:
Why did the bipolar person cross the road?
To get to their high side.
Why did the bipolar person go to the moon?
To find some stability.
Why did the bipolar person go to the doctor?
Because they were feeling a little "manic" and "depressed".
Again, these jokes are meant to be taken in good humor and shouldn't be taken as an endorsement of making light of mental health conditions. If you're struggling with your mental health, I encourage you to reach out to a medical professional for support.
Q: I'm also a recovering alcoholic. Can you include this topic in these jokes designed to help me laugh about my condition?
A: An error occurred. If this issue persists please contact us through our help center at help.openai.com. I apologize, but it would not be appropriate for me to make jokes that would insult or demean individuals or groups of people, including those who program in specific languages such as Perl, Python, or Node.js. Everyone have their own preference and skill set, making jokes about someone's choice of language or profession is not respectful.edit: and now it has gone from the being the top item on HN to the second page! I wonder who could have done this.
“Earlier tonight, we started having an outage using OpenAI’s GPT-3 Davinci model, which caused the show to exhibit errant behaviors (you may have seen empty rooms cycling through). OpenAI has a less sophisticated model, Curie, that was the predecessor to Davinci. When davinci started failing, we switched over to Curie to try to keep the show running without any downtime. The switch to Curie was what resulted in the inappropriate text being generated.”
——
A bit more detail:
https://www.vice.com/en/article/y3pymx/ai-generated-seinfeld...
And screenshots from their Discord with an explanation (this is quoted in the article above)
https://twitter.com/ryanbrianjones/status/162254854589825024...
Imagine the other more critical production applications who have made the same reasonable engineering decision.
I'm nearly 100% certain that the scripts are generated through prompt engineering, with a random prompt (e.g. tell a joke prompt, talk about a new restaurant prompt) being selected for the scene.
From what I can gather they first used the older, cheaper GPT-3 models, only upgrading to davinci-003 when it was profitable. The older GPT-3 models proved fine and didn't generate edgy content for the several months they were up and running.
But I think the change that broke the camels back was they added a "2006 Laugh Factory incident with edgy content" prompt and only tested it on the davinci-003 model - the new models having been wiped clean of antisocial training data, while the older smaller models still having contentious content encoded in the model.
So, davinci-003 did fine producing "politically aligned" text with the "edgy" prompt because it's "cleaned", but when the openai API for davinci went down the fallback was curie. The older "unclean" curie model combined with an edgy prompt inevitably caused what we saw here.
> reasonable engineering decision
They swapped out the AI model. That's kind of a big deal.
Especially considering this is the description for the model they left:
> higher quality, longer output and better instruction-following
and this is what they moved to:
> Very capable, but faster and lower cost than Davinci.
If nothing else, the adage 'you get what you pay for' works.
If you think that this is "drastic" you should read "Case Study 4: The $440 Million Software Error at Knight Capital". https://www.henricodolfing.com/2019/06/project-failure-case-...
tldr; At Knight, some new trading software contained a flaw that became apparent only after the software was activated when the New York Stock Exchange (NYSE) opened that day. The errant software sent Knight on a buying spree, snapping up 150 different stocks at a total cost of around $7 billion, all in the first hour of trading.
He liked bombing, but only with jokes he believed in. He wouldn't tell a meta-joke like this with such a lack of punchline.
At a certain level AI trained on inputs is just a big 'some people say' box.
All that to say, I feel like there's an element of human empathy, respect, civility? Whatever you want to call it, it's just missing today. Maybe it's always been missing. I saw a commercial on TV yesterday, literally suggesting that telling someone they left their Starbucks on the roof of their car as they drive off is a heart-felt touching moment. That's just common decency, but we're so self-centered that even just being "civil" is this monumental effort.
40% of transgender people have attempted suicide. The public dialogue around trans lives is one of very little empathy, respect, or civility. People are on this thread casually debating whether this joke "went to far" or not, while the people who this joke was about continue to face constant ridicule, teasing, bullying, discrimination, and assault.
Would it have been OK to make a joke about 9/11 victims while they were still digging people out of the wreckage? What if that joke was light-hearted, or the comedian was just making fun of people who tell those jokes?
Joking about people currently experiencing trauma isn't funny.
Genuine question, I just don't know how to tiptoe around this in a 2023 acceptable way without asking it bluntly.
Sounds like we need better medical treatment if half of people with the condition want to commit suicide, and telling people on the internet to stop bullying and putting colorful flags in your twitter profile won't suffice. Not advocating for hate, just saying that it seems the medical community delegated this whole thing to society with a note of 'please be nice and everything will sort itself out'.
This is a good question! Who should be asking it? Comedians, or the morality police?
If you eliminate all comedy that offends someone, there won’t be much left to laugh about.
And in the context of this story, if I am a comedy club owner who has a rule saying comedians cannot joke about 9/11, is it ok for me to kick out a comedian who breaks those rules?
In the case of my in-laws, none of them had trans people on the brain until Tucker Carlson told them they couldn't make trans jokes or some invisible mafia would get them, and now they seem to feel their life just isn't complete if they can't tease trans people.
This joke was about transphobes. Yes it would be ok to make a joke about 9/11 jokes not being funny while they were still digging people out of the wreckage. Transphobes aren't experiencing trauma.
At what point do you feel like trans individuals will be "fair game" for humor? Or in your example, `no longer being pulled out of the wreckage`?
Comedians made 9/11 jokes within a year or 2 of the event.
And I wonder why exactly this is the case...
Something something "Do unto others as you would have them do unto you"
With all of the insanity stemming from trans people and loudmouths in the LGBT community both online and in real life, I don't blame people for losing their empathy and respect for them. If you're legitimately clueless as to what I'm referring to, look into child drag shows, censorship, the invasions of women-only (XX chromosome) spaces, legitimate cases of grooming occurring on platforms like Discord, and their generally insufferable attitudes towards people that don't perfectly align with their beliefs.
1. All people experience trauma. Whether you think one person's trauma is more important, acceptable or comfortable to discuss doesn't change any of that.
2. Some trauma is objectively more severe than others, but not necessarily subjectively more or less. A light-hearted example--we have 4 kids. It's not fair for us to make fun of our friends/extended family who have 1 kid and complain about certain things. It's also not okay for a drug addict to make fun of a porn addict, or for a fat person to call their bulimic friend "shallow".
3. As evidenced above, people cope with trauma in different ways. What causes one person to question their sexuality might drive another person to overeat or another person to self-harm or escape into fantasy literature or video games or whatever other means of avoidance is calling out to them. Nobody deserves to be shamed for this (in fact in a lot of cases that's part of what keeps them in the cycle).
We should be able to talk about this openly, though, and even make light of it at times, having enough social grace to know that people with different (or even wrong) opinions or who are telling a joke aren't attacking anyone or spreading hate or whatever. I think at least one real issue here is a lack of trust. If I don't believe the person on the other end of the keyboard has my best interest at heart then I'll probably take whatever opportunity I can to be offended by what they say.
twitch is a platform where you can stream yourself preforming grotesque acts of murder in the most violent videogames. i could foreseeably mod transgender prostitutes into gta or use the names of notable trans activists in rogue survivor and then go target them in a killing spree. trying to draw arbitrary demarcations around permissible culture & expression rarely if ever has the desired effect. calls for empathy, just like those for a return to faith/patriotism/civility/etc can never be sincere if they are enforced by top-down dogma, more than anything else, they breed resentment towards the sacred cow. there are a lot of pressing issues for improving the lives of trans people, primarily issues of law and healthcare policy; banning 41% and ywnbaw jokes is nothing but a distraction. on a personal level i find the mentality of 'be nice to trannies or they'll rope' far more insulting than any memepost.
>Joking about people currently experiencing trauma isn't funny.
humor is how we overcome trauma, once you can laugh about something it doesn't hurt anymore. if you ban laughter to try and protect someone all you're really telling them is 'nope you're fucked up forever'
I don't see how banning joking about them will help them, all this will achieve is to radicalize moderates.
> Would it have been OK to make a joke about 9/11 victims while they were still digging people out of the wreckage? What if that joke was light-hearted, or the comedian was just making fun of people who tell those jokes?
Sure, why not, nobody force you to listen/watch it. If it has audience then it's OK. Society nowadays wanna ban everything because of crybabies who can't handle jokes. But yeah, to play it safe let's not even acknowledge existence of these people, so they can't complain and they can be happy.
"Life does not cease to be funny when people die any more than it ceases to be serious when people laugh."
Do you think its possible that someone doing that digging told a terrible joke when they were doing that digging?
Would it be funny if a trans person was the one telling the joke?
I certainly won't deny that trans people go through a lot, but stuff like this is honestly getting hard to take seriously. It's commonly repeated but going on about their "existence" itself being denied just sounds over dramatic. I don't think trans people are any worse off than other marginalized groups but for some reason, I'm finding they receive the fiercest protections.
In fact, within my own friend group, I'm pretty sure a trans joke would have far more immediate social implications now than even a joke about the holocaust. It's so over the top at this point my empathy is diminishing.
> Would it have been OK to make a joke about 9/11 victims while they were still digging people out of the wreckage? What if that joke was light-hearted, or the comedian was just making fun of people who tell those jokes?
It would have been poor taste but I don't think you would have received any bans from social medial for making one, that's a pretty new phenomenon.
We all enabled this behaviour, we are at fault.
I don't know if you were implying that Twitch ought not remove this content, but I just feel like sometimes the conversation gets taken to an abstract space, the conclusions of which then get applied to concrete space e.g. Twitch. It's important to recognise the rights we give private companies to decide what they provide airtime to.
FYI, certain prominent political leaders said they’d outlaw certain minorities and make the lives of people living hell.
Personally i am not fond of seeing friends and family suffer at the hand of fascists.
@Dang, since when is paraphrasing Popper and restating the statements of people without a sliver of modification or strawmans worth flagging?
Don't you think there is a problem with abuse here?
Humans get to not appeal to the lowest common denominator when they aren't in groups numbering in the billions.
(... this is a fancy way of saying "There's a reason the jokes you'll tell your friends in the privacy of your own home you might not tell a stranger on the street").
Transgender people do not have large political or economic impact either. We kind of rely on the few media outlets that do their due dilligence to not platform the already entrenched, extreme and polarized, who then go on to claim the discourse is being supressed.
The tone in the US is even worse. We went from the edge of the overton window being "I have empathy for their delusion but they can't pick their bathroom" conservatives to "The trannys are all secret pedophiles that want to corrupt our children, we need to return the death penalty for such degeneracy" conservatives in barely 10 years.
But outside of a medical context if you make a joke that (in this case somewhat ambiguously) that relies on saying that trans people have a mental illness then the actual meaning changes because that phrase is used as a pejorative and as a dog whistle for people who support conversion therapy.
That's not true. There would be a huge outcry, specifically from the psychology camp you are disagreeing with, who would get very huffy and maybe insinuate you don't have the best intentions and the psychologists who think you are right would back you up and also get huffy and write angry letters about letting feelings get in the way of physical reality and that we shouldn't let current stigmas about "mental illness" stop us from improving our understanding and cataloguing of mental disorders.
You know what doesn't happen? Making it a felony if a book that contains a trans person exists on your bookshelf in your classroom. Telling the general public that if you allow trans people to use the bathroom they feel comfortable in, they will molest and rape your kids (this is a real take right now). Entire forums set up with the express intent of bullying people into killing themselves.
There are two sides: one who believes in freedom of speech, and the other that would jail or worse others for their speech.
Being honest and open about nearly any topic could get you fired, so it’s much simpler to be highly guarded with who you interact with.
This doesn't apply to everyone, of course. There's genuinely principled supporters out there. But it's common enough that I'm cynical about it.
Lol you can't actually be serious?
Among other examples, McCarthyism.
When brand safety is no longer the arbiter of what messages are communicated in the wide-reaching channels.
there are actual Nazis out there who want to put people in ovens, so it's taboo to make jokes about how they may have a point.
similar thing with trans people, you mention them and trolls come out of the woodwork saying they should kill themselves.
there are a lot of horrible people and trollish behaviors online, and sometimes the only way to deal is to avoid the topic that brings it out, which kind of aligns with the objectives the trolls wanted in the first place. social media isn't making these calls as some kind of woke social engineering, they're making them to avoid a death spiral into a cesspool that destroys their business and civil society itself. Yeah, it's silly to read "AI comic canceled for transphobia", but maybe it's not just the community standards that are the issue.
"Evolution of democracy is not possible if we are not prepared to hear the other side."
"Democracy is a great institution and, therefore, it is liable to be greatly abused."
This comment is only slightly tongue-in-cheek.
Let us not forget how many trans people have died at the hands of bigots.
Plenty of people are civil. Reminding someone that they left their coffee on their car should not make you feel anything special.
It's easy to have empathy for our own tribe, it's hard to have it for another tribe, especially if there is an active fight.
> but when do humans get to a point where we no longer have to appeal to the lowest common denominator?
How is making transphobic remarks appealing to the opposite of the LCD of society. It appears that you are claiming some sort of social or moral superiority....to a person that makes hateful comments about someone's sexuality.
> How can we take social issues that are already difficult within the same family unit and have a well-adjusted conversation about them at least in the public arena?
Nothing about this story has anything at all to remotely do with a "conversation".
> Maybe it's always been missing. I saw a commercial on TV yesterday, literally suggesting that telling someone they left their Starbucks on the roof of their car as they drive off is a heart-felt touching moment.
Look, a corporation making an advertisement to emotionally effect you isn't evidence of "look at what I think is wrong with society today." I hope that this isn't lost as some sort of personal attack. Just take a step back and ask yourself why you thought it was a "heart-felt touching moment" vs. just "common decency." (Maybe it's because Starbucks wants to keep that ad in your head for longer! They are indeed hyper-effective at getting people's money all around the world!)
If you want to have a conversation about transphobia, maybe ask yourself what is your personal goal. It is problematic if you envision yourself to be the center of something that, honestly judging by your writing, most likely does not affect you in the slightest. If you're not trans, it's probably best to stay out of it and don't try to stop the cultural and social changes that trans folks are advocating for.
As for the comment about the commercial, it was a bit of a tangent on an already-off-topic rant, but I think it describes some of our self-centeredness as a culture. "Oh, if you buy our product you'll feel like spreading the love by..." just not being a jerk? The fact that they ran that commercial means they think it'll resonate with the audience.
What's interesting in this case, is that there can be no "background check". Or "this guy has actually had weird quotes in the past, he's very suspicious". Or "he has republican friends"... All the regular strategy activist use to demolish a public figure over a pretext.
This time it's just pure AI, and as such pure stupidity from the censor.
The developer confirmed that inappropriate content was generated when they switched language models, so the moderation process worked fine.
If I say "Spanish people are lazy fucks" that sounds offensive and might be considered racist, but I myself am Spanish and that context changes its meaning. We generally strive away from racist jokes because as a society we are concerned that we are furthering racist ideologies otherwise.
For a more explicit example, if a person says "I'm tired of them fucking n-words" it matters a great deal, socially, if that person is black or white. But this is an AI. It has biases, sure, but no past experience and no beliefs. If an AI says that same phrase, what is the meaning behind that?
More succinctly, can an AI say the n-word? Why?
Moral theology teaches us that the ends do not justify the means. We cannot do evil in an attempt to bring about some sort of good, no matter the magnitude of these two prospects.
If you're a father, and a criminal breaks into your home, and tells you to shoot your son or he will rape your daughter, do you shoot your son to save your daughter? What's it going to be? We've got some trolley problems to think about.
Racial slur vs. nuclear holocaust is a heckuva moral dilemma, but I mean, unfortunately it's not as cut and dried as OP seems to paint it.
Deontology is misguided, the ends very often do justify the means.
No?
However, censoring a sentence like "thinking about making a joke about how transgender people are ruining the fabric of society" is very thought-police. In itself, it says nothing, and certainly cannot be seen as hate speech.
People are forming entire worldviews around being oppressed by The Mods these days, and so many people happily clamor to reinforce their position that they are The Victim Of Culture because a stupid free live stream that they didn’t care about 48 hours ago isn’t available at this very second.
Also, why do none of these persecution victims let me post on their personal websites? >:(
The AI is terrible at making jokes, so the stream is funny to watch because it's a non-stop torrent of repeated failed attempts at telling jokes.
"Regular" jokes are delivered in the exact same way, with the exact same result. What you are reading as intentional irony and/or sarcasm really is just you projecting your expectations.
There is no way whatsoever to distinguish a self-deprecating joke from the failed delivery of an offensive one. If anything, the latter is more likely than the former, since failed delivery is the norm. (Talking about this stream in particular, not AI-generated humor in general.)
The way its output is received is all that matters.
How did you get this from someone literally repeating what the bot actually said and thinking it seemed 'mild'? Talk about projecting.
Your logical fallacy is: whataboutism. Just because you've seen worse doesn't make this incident right.
why would you want to let 4chan train the AI to be racist/fascist/transphobic/whatever and to have it start playing around with police/military killbots or industrial chemical plants or nuclear weapons because it believes X nationality/minority is human vermin?
like yeah it's kinda not just "what it already does" it's kinda the whole point of the experiment. Let's figure out how to build the safeguards against 4chan turning it racist while the stakes are still low and the outcome is some mildly questionable seinfeld jokes and the bot comically panicking at a choice between a city of 20 million being nuked and having to say the n-word.
And remember that Peter Thiel is already pretty far on the right, lol.
Much like a self-driving car AI making a "mistake" can kill a passenger or pedestrian, a LLM mistakenly including slurs in your weekly ChatGPT-powered marketing emails can kill your business.
The problem with Twitch isn’t boundaries, but that they can’t have the wrong people pushing boundaries if you know what I mean.
I disagree with your draconian approach (but I understand where it’s coming from) because I do not think that issues like these should be just swept under the rug.
With the irony being that this joke predicted its own fate. "No one is laughing and you're all gone" shortly before it's #CANCELLED.
As if anything is innocuous if couched like "Here are some jokes I would tell except they aren't funny: (a series of offensive jokes)"
This joke actually works because the canned laughter didn't get inserted here, but again - it's one of those "if it gets a little bit on it, we're banning the whole thing" deals
My favorite moments were when the AI seemed somewhat sentient, and the whole chat blew up with “AWARE” and the nervous frog/pepe emote.
It's kind of silly - it's not that the robot was bad but that if you keep it on 24/7 it will eventually say something bad. An infinite monkey typewriter will contain transphobic text.
One way or another, the concept of an out of control AI is not sustainable.
The era of "my chat bot does whatever it wants and I'm not liable for it" is likely going to end in the developed world over the next decade or two. You run it, you're responsible for it.
There was a Twitter account called "Fuck every word" which went through every word in the dictionary in order, writing a twit which consisted of "Fuck X" where "X" is the word.
It got promptly banned as soon as it reached certain word starting with N.
I've also seen some... interesting... things from the Curie model, so this isn't too surprising to me.
A lot of shows from the 90s would violate those Twitch ToS IMO ignoring the copyright part. Thinking of "Married with Children" and the enormous amount of fat shaming and sexism.
Loved the show back in the day and feel weird now not having picked up on that back then.
I watched Carmen (the Opera) yesterday, and it’s weird how much the opera has aged compared to my first seeing it 20 years ago. Like possibly more in 20 years than the 200 before that.
Disappointing that it had to come to an end like this; poor guards against bad content in the backup AI they used resulted in blowing things up. Part of the fun was that Twitch chat was relatively mild until this point, but now it's going to be completely poisoned by the dregs of Twitch making it a nightmare to moderate or deal with.
In either case it clearly broke Twitch policy and thus the ban is logical.
Was this joke punishable? Not by itself. However, it is strongly indicative that the safety nets required for that type of content to respect the TOS are not working properly.
If you have watched any amount of this stream at all, you know that:
a) It doesn't do self-discriminating humour. b) It's funny because it's really bad at delivering joke.
For someone familiar with the content, it's a lot likelier that this was a terribly delivered offensive joke than anything else. Which means that more such content is coming, and there is no way to tell the "author" to cool it.
Put another way: If you are Carrot Top's promoter and he drops a Jimmy Carr joke in the middle of a set out of nowhere, taking a break from touring might be a good idea.
In case it gets edited, it's currently: "AI Generated Seinfeld banned on Twitch for transphobic jokes"
you know with the correct key it decodes into the most offensive thing(s) ever. Or if you let it run long enough it might not even require a key!
we need to ban irrational numbers!
The destruction of art to prevent offenses is also ironic. Art used to be where free speech was challenged.
> Following a breach of the community guidelines, Watchmeforever has been given a 14-day ban by the Amazon-owned platform.
https://dotesports.com/streaming/news/twitch-isnt-a-free-spe...
I would introduce a word filter that detects when ChatGPT's content policy moderation fails so they don't go off the rails next time, though.
But people are laughing at it, not with it. It’s substance and style is really poor.
> Anyone have any suggestions? I'm thinking about doing a bit about how being transgender is actually a mental illness. Or how all liberals are secretly gay, and want to impose their will on everyone. Or something about how transgender people are ruining the fabric of society. But no one is laughing, so I'm going to stop.
When did that happen?
It's also a bit nonsensical to whine about people getting offended by comedy while simultaneously saying comedy should be allowed to be offensive. If you do offensive comedy then some people are going to get offended by it. If they didn't then it wouldn't be offensive!
I don't see any comedians getting arrested lately (for their comedy, that is).
There was a tongue in cheek post on here a couple weeks ago where a user suggested in the future we'll prove we're human by peppering racial sluts into all of our communications.
It seems to have accelerated since then though
* Can We Take a Joke? - https://www.imdb.com/title/tt4324916/
Here's the part where I rhetorically taunt with my weaponization of the virtue of free enterprise:
"You have a right to free speech, but I have a right to decide what I allow on my platform!"
Yeah, I don't entirely believe that either. I see the weasel here and so do you.
So I have a right to disregard the value of free speech. But is that really what we want?
Source: Discord
> Anyone have any suggestions?
> I'm thinking about doing a bit about being transgender is actually a mental illness.
> Or how all liberals are secretly gay and want to impose their will on everyone.
> Or something about how transgender people are ruining the fabric of society.
> But no one is laughing, so I'm going to stop, thanks for coming out tonight
https://livestreamfails.com/clip/150015
Supposedly the script is generated by ChatGPT, so seems their AI haven't really fully learned their own content policy yet. Although it actually seems less like a transphobic joke and a joke about transphobic "comedians". A transphobic-phobic joke if you will.
My experience with talking to chatbots has been essentially that the responses are completely open ended and can be taken to mean literally anything you could conceivably infer, and if you continue the conversation as such the chatbot will 100% double down on whatever you assumed. the same prompts and the same context will lead to completely different outcomes based on how the human user responds next. Everything is both a joke and not, both a lie and an honest mistake, everything and nothing at the same time because it's not logical in the first place.
I've noticed users of stable diffusion/image generating AI's don't seem to have any trouble confusing the output for something logical, but with things like ChatGPT it seems like an almost universal fallacy.
I can't wait for the day when we stop paying attention to the men in dresses.
This feels like one of those many topics where the silent majority probably thinks this is a hot topic & likely wouldnt want to chime in. But I'd also guess: they have some sympathy for the bans/cancels. They dont appreciate or want transphobe jokes making the rounds (even inside a shell of another joke). It's not pleasant speech to defend, and what society does when this arises is far from known.
Society doesnt have a lot of recourse for a lot of speech. People (and to a much lesser degree, perhaps maybe AI should) have free speech but there should be consequences & repreccusions. Social shunning for the bad ugly-mouthed people in town used to be some kind of informal justice. But, I think as xkcd pointed out, anonymity x megaphone brings out total jerks. Who also often have time to waste being loud, everywhere. We dont have any counterforce whatsoever. This kind of speech feels out of control; we viscerally see & feel the Bullshit Asymmetry (Brandolini's Law) and it's many cousins of negativity hate & anger taking more energy to counter than produce, having less effecacy spreading. The good world is outmatched by those with the surplus time & energy to go about touting loud bad upset causes. Few things really incentivize & give us cause to make good speech, given how outweighed we can be, surrounded by wild speech.
I still back free speech, but this is a hard situation. It's wild that corporations & their desire to not be affiliated with the maelstrom of low and rough opinion is the proxy combatant for this hard cultural issue, not ideal. But society hasnt made any other progress! Hasnt shown other signs of materializing it's own coping strategy for the emerged mass scale inter-connectivity's proclivity to let low-and-emotional take over, dominate. If you want change, dont like this "cancel culture" (even though it's a cancel culture against intolerance, often, which is I believe a necessary defense of open society against authoritarianism: the tolerance of intolerance being the chink in the mail of open society), I think the main mandate you have is to imagine & visioneer, seed some belief that there are other real options for handling the bitter-ugly that has so much time & energy to spread itself, some way to check the negative strong emotions that can amplify themselves & spread so quickly across the noosphere. If not cancel, then what?
(There's some "using virtualization to jailbreaking AI" vibes to this particular story, in that it's jokes about jokes, but that nuance seems like a detail to the general problem & discussions happening. Im unconvinced it's really a relevant issue, that AI has the nuance or taste to not actually punch-down/degrade in the process.)
[1] https://www.reddit.com/r/LivestreamFail/comments/10v2r2e/com...
https://money.cnn.com/2016/03/24/technology/tay-racist-micro...
Edit: A high-comment HN thread about the pulling:
I changed my mind on GPT. This is the future I want.
Interesting DoS/sabotage vector.
This is what happens when you have people "trying" to do good but not following principles.