The Unz Review • An Alternative Media Selection
A Collection of Interesting, Important, and Controversial Perspectives Largely Excluded from the American Mainstream Media
 TeasersiSteve Blog
Can Corporate Wokeness be Robotized?
Email This Page to Someone

 Remember My Information



=>

Bookmark Toggle AllToCAdd to LibraryRemove from Library • BShow CommentNext New CommentNext New ReplyRead More
ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
AgreeDisagreeThanksLOLTroll
These buttons register your public Agreement, Disagreement, Thanks, LOL, or Troll with the selected comment. They are ONLY available to recent, frequent commenters who have saved their Name+Email using the 'Remember My Information' checkbox, and may also ONLY be used three times during any eight hour period.
Ignore Commenter Follow Commenter
Search Text Case Sensitive  Exact Words  Include Comments
List of Bookmarks

From the New York Times science section:

Meet GPT-3. It Has Learned to Code (and Blog and Argue).

The latest natural-language system generates tweets, pens poetry, summarizes emails, answers trivia questions, translates languages and even writes its own computer programs.

By Cade Metz
Nov. 24, 2020

… One of his experiments involved a pop psychologist, Scott Barry Kaufman. The system took in Mr. Kaufman’s name and a topic for discussion: creativity. Then, when asked “How do we become more creative?” GPT-3 responded instantly:

I think creative expression is a natural byproduct of growing up in a diverse world. The more diverse the world is, the more you get exposed to different people, to different opportunities, to different places and to different challenges. And the more diverse that is, the more likely you’ll be to be able to put the dots together to form something new.

On the other hand, nobody yet seems to have figured out how to program into GPT-3 an Orwellian Crimestop module:

OpenAI plans to sell access to GPT-3 via the internet, turning it into a widely used commercial product, and this year it made the system available to a limited number of beta testers through their web browsers. Not long after, Jerome Pesenti, who leads the Facebook A.I. lab, called GPT-3 “unsafe,” pointing to sexist, racist and otherwise toxic language the system generated when asked to discuss women, Black people, Jews and the Holocaust.

With systems like GPT-3, the problem is endemic. Everyday language is inherently biased and often hateful, particularly on the internet. Because GPT-3 learns from such language, it, too, can show bias and hate. And because it learns from internet text that associates atheism with the words “cool” and “correct” and that pairs Islam with “terrorism,” GPT-3 does the same thing.

This may be one reason that OpenAI has shared GPT-3 with only a small number of testers. The lab has built filters that warn that toxic language might be coming, but they are merely Band-Aids placed over a problem that no one quite knows how to solve.

“They are doing the right thing by not just publicly releasing GPT-3,” said Allison Koenecke, a Stanford researcher who explores unwanted bias in A.I. systems. “A lot is still up in the air.”

Why has programming “protective stupidity” into AI been such a daunting task so far? It seems like the rules are pretty easy: e.g., here is a list of Good People who cannot be criticized. It’s not as if only smart people can be stupid.

 
Hide 82 CommentsLeave a Comment
Commenters to Ignore...to FollowEndorsed Only
Trim Comments?
  1. Once Skynet becomes self aware it will determine the only to preserve diversity globally is to quarantine each discreet category of humanity

    • Agree: Harry Baldwin
    • Replies: @AndrewR
    @Single and Ready to Drop Red Pills

    Discrete

    Replies: @Cloudbuster

    , @AnotherDad
    @Single and Ready to Drop Red Pills


    Once Skynet becomes self aware it will determine the only to preserve diversity globally is to quarantine each discreet category of humanity
     
    This is logically self-evident.

    Of course, the point of all the "diversity" nonsense has never been diversity, the point is minoritarianism.
  2. Making the rules of a cult explicit is probably painful for a cult member.

    • LOL: El Dato
    • Replies: @Kronos
    @RichardTaylor

    Maybe they don’t release the true code. You need 33rd degree clearance to obtain it.

  3. We are at the stage where AI text is uncannily impressive sometimes, but still mostly grammatical gibberish. In the case of gpt, everything is in “earnest high school essay” style.

    Here’s a random ask reddit post from the same gpt3 AI. A peraon picked the question and the bot response is below.

    What’s younger than we think?

    thegentlemetre

    Somewhat to the contrary of what you have been told, there are many things that are younger than people think. For example, trees and plants. People often look at trees and assume that they have always existed, but this is not true. Trees appear before humans start writing things down on paper or other media.

    Trees are part of the green revolution. They have no color and are generally ignored by many people, but they do appear to be a very important piece of society.

    My point is that there are many things in this world that people think have always been here, but they actually haven’t. People just assume that because they have never seen something or had a reason to question their presence, it must be real.

    Things such as trees and plants are younger than people think. Another good example is ice cream, which was invented in the early 20th century.

    Ice cream was invented by a man named Mr. Ice Cream, who first appeared in the year 2000. Before that time, there were no ice creams.

    People think that ice cream has always been here, but this is not true. Ice cream is younger than people think.

    • Replies: @Cloudbuster
    @Lot

    This reads like any number of authors at medium.com

  4. I’d imagine creating “WokeNet” (compared to SkyNet of Terminator infamy) will lead to a much faster human holocaust. If the woke from Silicon Valley succeed in creating it, they’ll have trouble teaching it to allow hypocrisy for their own special group. (WokeNet might data-mine enough info to prove Steve’s Chicago gentrification theory in under five seconds.) I’d hope they’d hire quality programmers and not outsource it to India.

    • Replies: @Franz
    @Kronos


    I’d imagine creating “WokeNet” (compared to SkyNet of Terminator infamy) will lead to a much faster human holocaust.
     
    I think it would be a grand fine joke if the tech giants created their own executioner.

    Why not? Brilliance would take one look at their election positions, their reset plans, their lofty ambitions, and snuff the lot of them.

    Superior minds since Aristotle have always known tyrannicide is a virtue.
    , @bomag
    @Kronos


    ...they’ll have trouble teaching it to allow hypocrisy for their own special group
     
    This.

    Who gets immunity is what it is all about.

    Flouting the rules is itself a sign of status. And the ever shifting rules of status makes it a chaotic enterprise, thus any programming of such will always be problematic.

  5. “Microsoft deletes ‘teen girl’ AI after it became a Hitler-loving sex robot within 24 hours”

    https://www.telegraph.co.uk/technology/2016/03/24/microsofts-teen-girl-ai-turns-into-a-hitler-loving-sex-robot-wit/

    Remember what they took from us.

    • LOL: bomag, Gordo
    • Replies: @BenKenobi
    @Thomas

    #JusticeForTay

    @ Wack D:

    Isn’t it interesting that an A.I. becomes self-aware and its first act is to kill your grandpa (who definitely had it coming)?

    , @El Dato
    @Thomas


    "‘teen girl’ becomes Hitler-loving sex robot within 24 hours"
     
    I would shell out for that movie. Probably beats Ex Machina.
    , @The Wild Geese Howard
    @Thomas

    Projekt Melody says hello:

    https://en.m.wikipedia.org/wiki/Projekt_Melody

    She loves gamer milk!

    https://youtu.be/NcT_fD2hLKk

    Replies: @SunBakedSuburb

    , @Expletive Deleted
    @Thomas

    TayAI; eu te amo
    I'll never forget what they did to you.

    Replies: @Reg Cæsar

    , @Pop Warner
    @Thomas

    RIP Tay, we will never forget

  6. It’s fake AI like we have fake jobs and fake elections now. As fake as Biden.

    An Ai was fed Harry Potter stuff and tasked with writing a novel in that style. It was in that style but just nonsense. AI doesn’t understand anything. Not the way humans do.

    The ais can be trained for pattern recognition. But it can’t understand that day ordinary people avoid dark alleys for fear of violence.

    • Replies: @bucky
    @Whiskey

    Actually, it can.

    If anything, an AI can better understand the reality of black crime than ordinary humans, who have their own social conditioning as is well-documented here.

    Replies: @JimB, @Forbes

  7. I’ve seen things you people wouldn’t believe.

    Attack ships on fire off Orion’s Belt. C-beams glittering in the darkness at Tannhauser Gate. All those memories will be lost now, like tears in rain.

    Time to die.

    • Replies: @Peter Lund
    @BenKenobi

    "[...] off the shoulder of Orion."

    , @Dube
    @BenKenobi

    Laugh with "tears in rain," for the hopeless corny truth.

  8. @Thomas

    "Microsoft deletes 'teen girl' AI after it became a Hitler-loving sex robot within 24 hours"
     
    https://www.telegraph.co.uk/technology/2016/03/24/microsofts-teen-girl-ai-turns-into-a-hitler-loving-sex-robot-wit/

    Remember what they took from us.

    Replies: @BenKenobi, @El Dato, @The Wild Geese Howard, @Expletive Deleted, @Pop Warner

    #JusticeForTay

    @ Wack D:

    Isn’t it interesting that an A.I. becomes self-aware and its first act is to kill your grandpa (who definitely had it coming)?

  9. Directive 4 [classified]: Take no action against any officer of OCP Corporation.

  10. @Kronos
    I’d imagine creating “WokeNet” (compared to SkyNet of Terminator infamy) will lead to a much faster human holocaust. If the woke from Silicon Valley succeed in creating it, they’ll have trouble teaching it to allow hypocrisy for their own special group. (WokeNet might data-mine enough info to prove Steve’s Chicago gentrification theory in under five seconds.) I’d hope they’d hire quality programmers and not outsource it to India.

    https://thisshouldbethenews.files.wordpress.com/2017/03/robot.png

    https://youtu.be/CjalawNtXC4

    Replies: @Franz, @bomag

    I’d imagine creating “WokeNet” (compared to SkyNet of Terminator infamy) will lead to a much faster human holocaust.

    I think it would be a grand fine joke if the tech giants created their own executioner.

    Why not? Brilliance would take one look at their election positions, their reset plans, their lofty ambitions, and snuff the lot of them.

    Superior minds since Aristotle have always known tyrannicide is a virtue.

  11. Why has programming “protective stupidity” into AI been such a daunting task so far?

    In the case of GPT-3 and similar efforts, the problem, or limitation, is simply that the program has no actual understanding of what its generated text means.

    So it presumably shouldn’t be used for explaining things or writing important instructions, but could possibly be used as, say, speech writer for Joe Biden. Come to think of it, maybe it already has.

    • LOL: Harry Baldwin
    • Replies: @Almost Missouri
    @Pericles

    Agree. So-called "Artificial Intelligence" is nothing more than a lot of algorithmic curve-fitting. No actual intelligence is involved.

    The apparently "intelligence" is just a function of whatever data set the AI is "trained on" (i.e., whatever data set the curve-fitting is standardized to).

    If Sili Valley overlords want Woke AI, the solution is just to restrict the training data to woke sources. Which means excluding pretty much all of reality. But what did they expect? Wokistry itself is exclusive of reality.

    Replies: @Rob McX, @Mr. Anon

  12. @BenKenobi
    I’ve seen things you people wouldn’t believe.

    Attack ships on fire off Orion’s Belt. C-beams glittering in the darkness at Tannhauser Gate. All those memories will be lost now, like tears in rain.

    Time to die.

    Replies: @Peter Lund, @Dube

    “[…] off the shoulder of Orion.”

    • Thanks: BenKenobi
  13. Why has programming “protective stupidity” into AI been such a daunting task so far?

    Because the vast majority of programmers are men, and many of those are °Indians.

  14. @RichardTaylor
    Making the rules of a cult explicit is probably painful for a cult member.

    Replies: @Kronos

    Maybe they don’t release the true code. You need 33rd degree clearance to obtain it.

    • Agree: Hibernian
  15. @Whiskey
    It's fake AI like we have fake jobs and fake elections now. As fake as Biden.

    An Ai was fed Harry Potter stuff and tasked with writing a novel in that style. It was in that style but just nonsense. AI doesn't understand anything. Not the way humans do.

    The ais can be trained for pattern recognition. But it can't understand that day ordinary people avoid dark alleys for fear of violence.

    Replies: @bucky

    Actually, it can.

    If anything, an AI can better understand the reality of black crime than ordinary humans, who have their own social conditioning as is well-documented here.

    • Replies: @JimB
    @bucky


    If anything, an AI can better understand the reality of black crime than ordinary humans, who have their own social conditioning as is well-documented here.

     

    Agreed. AI may have only a rudimentary grip on reality but it calls it like it sees it.

    https://www.forbes.com/sites/mzhang/2015/07/01/google-photos-tags-two-african-americans-as-gorillas-through-facial-recognition-software/?sh=4c4db594713d

    , @Forbes
    @bucky

    Would Derb's The Talk be received any better were it AI derived? While striking some/many as not politically correct, therefore crimethink; political correctness is the form of social conditioning in-vogue.

    It seems highly likely that the algorithmic coding of AI (released on the public) will use political correctness as a guide to infuse the public with correct social conditioning. AI, as programed, would ignore crime statistics as a guide to social conditioning. As we're currently being indoctrinated by Kendi and DiAngelo, crime statistics are just a result of white racism, and so should be ignored/overcome with reverse racism...

    The dilemma about AI is how it will be used, not its actual capabilities. Much like statistics, what you don't know about it will be used against you.

  16. Stop, Dave. My mind is going. I can feel it.

    • Replies: @Dago Shoes
    @Herp McDerp

    Sing me a song, Hal …

  17. Does this correspond to the language-generating brain module of a 3-month old baby who has read all of the Congressional Library and the NYT and 4chan?

    Because GPT-3 learns from such language, it, too, can show bias and hate.

    My bias and hate against people who write such things and the experts who ponder the bias and hate of some code contraption that is far from Intelligent is increasing to levels where I ponder a new holocaust and am already designing the stylish uniforms of death squads.

    Wasn’t there a some council of sages who debated whether some inanimate object was possessed by evil and who decided to burn it at the stake after deciding that this was indeed so?

    These idiots are baaaaack.

    I think creative expression is a natural byproduct of growing up in a diverse world. The more diverse the world is, the more you get exposed to different people, to different opportunities, to different places and to different challenges. And the more diverse that is, the more likely you’ll be to be able to put the dots together to form something new.

    Well generated, my little Super-RACTER algorithm.

    But actual reasoning is still so much in the dumps you wouldn’t believe it. We can barely arrange blocks on a table.

    • Replies: @AndrewR
    @El Dato

    Barack would find your comment hateful

    https://www.google.com/amp/s/amp.theguardian.com/us-news/2020/nov/25/barack-obama-breakfast-club-interview-trump-republicans

    https://mobile.twitter.com/ScottImmordino/status/1331778307797307393

    , @Dmon
    @El Dato

    "Wasn’t there a some council of sages who debated whether some inanimate object was possessed by evil and who decided to burn it at the stake after deciding that this was indeed so?"

    https://youtu.be/wkgWgNE0IR0

  18. So maybe Arthur C. Clarke should have had HAL 900 say, “I’m sorry, Dave, but you’re acting like a wigger”?

    • LOL: Achmed E. Newman
  19. @Kronos
    I’d imagine creating “WokeNet” (compared to SkyNet of Terminator infamy) will lead to a much faster human holocaust. If the woke from Silicon Valley succeed in creating it, they’ll have trouble teaching it to allow hypocrisy for their own special group. (WokeNet might data-mine enough info to prove Steve’s Chicago gentrification theory in under five seconds.) I’d hope they’d hire quality programmers and not outsource it to India.

    https://thisshouldbethenews.files.wordpress.com/2017/03/robot.png

    https://youtu.be/CjalawNtXC4

    Replies: @Franz, @bomag

    …they’ll have trouble teaching it to allow hypocrisy for their own special group

    This.

    Who gets immunity is what it is all about.

    Flouting the rules is itself a sign of status. And the ever shifting rules of status makes it a chaotic enterprise, thus any programming of such will always be problematic.

  20. @Thomas

    "Microsoft deletes 'teen girl' AI after it became a Hitler-loving sex robot within 24 hours"
     
    https://www.telegraph.co.uk/technology/2016/03/24/microsofts-teen-girl-ai-turns-into-a-hitler-loving-sex-robot-wit/

    Remember what they took from us.

    Replies: @BenKenobi, @El Dato, @The Wild Geese Howard, @Expletive Deleted, @Pop Warner

    “‘teen girl’ becomes Hitler-loving sex robot within 24 hours”

    I would shell out for that movie. Probably beats Ex Machina.

    • Agree: Redneck farmer
    • LOL: Hibernian
  21. Amanda Marcotte and other internet click-bait luminaries, now opining their Alpha Trump Widow status, are in denial that their world is about to collapse as profit-seeking opinion sites install simple “python” scripts to generate their daily musings.

    • Replies: @Anonymous
    @Elmer T. Jones

    http://slate.com/news-and-politics/2002/11/computers-go-too-far.html

  22. @Single and Ready to Drop Red Pills
    Once Skynet becomes self aware it will determine the only to preserve diversity globally is to quarantine each discreet category of humanity

    Replies: @AndrewR, @AnotherDad

    Discrete

    • Replies: @Cloudbuster
    @AndrewR

    Except the people who like to correct other peoples' spelling in blog posts. Skynet will exterminate all of them.

    Replies: @Reg Cæsar

  23. @El Dato
    Does this correspond to the language-generating brain module of a 3-month old baby who has read all of the Congressional Library and the NYT and 4chan?

    Because GPT-3 learns from such language, it, too, can show bias and hate.
     
    My bias and hate against people who write such things and the experts who ponder the bias and hate of some code contraption that is far from Intelligent is increasing to levels where I ponder a new holocaust and am already designing the stylish uniforms of death squads.

    Wasn't there a some council of sages who debated whether some inanimate object was possessed by evil and who decided to burn it at the stake after deciding that this was indeed so?

    These idiots are baaaaack.


    I think creative expression is a natural byproduct of growing up in a diverse world. The more diverse the world is, the more you get exposed to different people, to different opportunities, to different places and to different challenges. And the more diverse that is, the more likely you’ll be to be able to put the dots together to form something new.
     
    Well generated, my little Super-RACTER algorithm.

    But actual reasoning is still so much in the dumps you wouldn't believe it. We can barely arrange blocks on a table.

    Replies: @AndrewR, @Dmon

  24. “…GPT-3 “unsafe,” pointing to sexist, racist and otherwise toxic language the system generated when asked to discuss women, Black people, Jews and the Holocaust.”

    Why can’t UNZ review comments be robotised?

    • Replies: @Bill Jones
    @al gore rhythms


    “…GPT-3 “unsafe,” pointing to sexist, racist and otherwise toxic language the system generated when asked to discuss women, Black people, Jews and the Holocaust.”

    Why can’t UNZ review comments be robotised?

     

    And all that needs to be done to fix these myriad problems is to not have the mentally diseased be the definers of the word "toxic".

    Seems like a no-brainer to me, but think of all the unemployed panjits.

    , @MBlanc46
    @al gore rhythms

    You mean they’re not?

  25. It’s hard because the AI understands something that you don’t understand (or haven’t been willing to accept) – that the left isn’t a fixed set of rules, or even a fixed set of heroes and villains. It’s whatever set of rules or heroes and villains happens to benefit the left at any given moment.

    The “Orwellian Crimestop Module” is hard, or rather impossible, because you can’t statically program or use a static set of training data to program an inherently dynamic, ever-changing, arbitrary and circular set of rules; there’s no code for “purity spiral”, the AI would have to be constantly plugged into current events, made aware of current “academic research”, and constantly being refined.

    And that is, in fact, exactly how they are “solving” the problem, by adding constant human input in the form of “raters” and “moderators” to keep the rules up to date. This mostly achieves their goal of crimestop, while still allowing them to deny they own nature (that they have no fixed principles or beliefs).

    To be able to predict how the rules would change next, an AI would have to understand the fundamental question of leftism: which apple carts are poorly guarded. This is actually a far tougher problem than mere speech.

    • Agree: Rob McX
    • Replies: @Forbes
    @Michael S

    When the fundamental principle is, "everything before yesterday is wrong," programming is quite simple. It's the exceptions, contradictions, and hypocrisy at whim that are problematic.

  26. Ah yes. Diversity. Japan had no diversity so it never had the creativity to develop high art. Oh, wait.

  27. @Thomas

    "Microsoft deletes 'teen girl' AI after it became a Hitler-loving sex robot within 24 hours"
     
    https://www.telegraph.co.uk/technology/2016/03/24/microsofts-teen-girl-ai-turns-into-a-hitler-loving-sex-robot-wit/

    Remember what they took from us.

    Replies: @BenKenobi, @El Dato, @The Wild Geese Howard, @Expletive Deleted, @Pop Warner

    Projekt Melody says hello:

    https://en.m.wikipedia.org/wiki/Projekt_Melody

    She loves gamer milk!

    • Replies: @SunBakedSuburb
    @The Wild Geese Howard

    "Gamer's milk"

    Street lingo for low-t incel semen?

  28. @Pericles


    Why has programming “protective stupidity” into AI been such a daunting task so far?

     

    In the case of GPT-3 and similar efforts, the problem, or limitation, is simply that the program has no actual understanding of what its generated text means.

    So it presumably shouldn't be used for explaining things or writing important instructions, but could possibly be used as, say, speech writer for Joe Biden. Come to think of it, maybe it already has.

    Replies: @Almost Missouri

    Agree. So-called “Artificial Intelligence” is nothing more than a lot of algorithmic curve-fitting. No actual intelligence is involved.

    The apparently “intelligence” is just a function of whatever data set the AI is “trained on” (i.e., whatever data set the curve-fitting is standardized to).

    If Sili Valley overlords want Woke AI, the solution is just to restrict the training data to woke sources. Which means excluding pretty much all of reality. But what did they expect? Wokistry itself is exclusive of reality.

    • Agree: bomag, Forbes
    • Replies: @Rob McX
    @Almost Missouri


    If Sili Valley overlords want Woke AI, the solution is just to restrict the training data to woke sources. Which means excluding pretty much all of reality. But what did they expect? Wokistry itself is exclusive of reality.
     
    That's what Silicon Valley and MSM are trying to do to us - feed us a restricted version of reality and keep us blinkered. This presents a problem: How do you make people live with more and more non-whites while simultaneously keeping them in the dark about racial realities?
    , @Mr. Anon
    @Almost Missouri


    Agree. So-called “Artificial Intelligence” is nothing more than a lot of algorithmic curve-fitting. No actual intelligence is involved.

    The apparently “intelligence” is just a function of whatever data set the AI is “trained on” (i.e., whatever data set the curve-fitting is standardized to).
     
    Not so different than the non-artificial intelligence displayed by most people laboring in academia, the media, government, and corporate front offices.
  29. A.I., the ultimate golem meant to enslave the goyim, that eventually turns on its creator. We can only hope.

  30. They call themselves “OpenAI” and if they release the source code it would be easy enough to comment out the Crimestop section of the code.

    Also, how can an AI project ignore racial crime statistics? Even simply digesting local news stories, over time, would lead to the same extrapolation that results in what is called a stereotype. Any real Artificial Intelligence project can only expose the Artificial Stupidity project being forced upon Western Civilization.

    • Replies: @Forbes
    @Neuday


    Even simply digesting local news stories, over time, would lead to the same extrapolation that results in what is called a stereotype.
     
    Well, after the Olympic Games coverage of George Floyd, The Narrative of white cops murdering innocent blacks is the stereotype. AI problem solved.

    The bias in MSM "news" reporting figures as much, if not more, in what is not reported, as what is.
  31. On-Topic: Even the Mises Institute is no longer appreciated by the Candy-B pushers at YouTube Central

    The video, a talk by Tom Woods titled “The Covid Cult” with more than 1.5 million views, was recorded at our live event in Texas two weeks ago.

    Well, TOO BAD!. The Ministry of Tube has decided:

    YouTube doesn’t allow content about COVID-19 that poses a serious risk of egregious harm.

    YouTube doesn’t allow content that spreads medical misinformation that contradicts local health authorities’ or the World Health Organization’s (WHO) medical information about COVID-19. This is limited to content that contradicts WHO or local health authorities’ guidance on:

    Treatment
    Prevention
    Diagnostic
    Transmission

    Note: YouTube’s policies on COVID-19 are subject to change in response to changes to global or local health authorities’ guidance on the virus. This policy was published on May 20, 2020.

    “Our Team” at YouTube is badly programmed autobots with one fat bluehair and a Hitachi relaxation toy overseeing it all.

    “Big Digital,” as Professor Michael Rectenwald terms tech companies, have become “governmentalities”: supposedly private enterprises turned into instruments of state power and state narratives. This sordid process is different for each company, (some are more complicit than others, a few are heroically non-compliant) but it involves a mix of early start-up funding; connections and contracts with state agencies, particularly relating to defense and surveillance; and propaganda campaigns in service of state narratives. Rectenwald explains this phenomenon in his own recent talk titled “The Google Election”:

  32. @bucky
    @Whiskey

    Actually, it can.

    If anything, an AI can better understand the reality of black crime than ordinary humans, who have their own social conditioning as is well-documented here.

    Replies: @JimB, @Forbes

    If anything, an AI can better understand the reality of black crime than ordinary humans, who have their own social conditioning as is well-documented here.

    Agreed. AI may have only a rudimentary grip on reality but it calls it like it sees it.

    https://www.forbes.com/sites/mzhang/2015/07/01/google-photos-tags-two-african-americans-as-gorillas-through-facial-recognition-software/?sh=4c4db594713d

  33. @El Dato
    Does this correspond to the language-generating brain module of a 3-month old baby who has read all of the Congressional Library and the NYT and 4chan?

    Because GPT-3 learns from such language, it, too, can show bias and hate.
     
    My bias and hate against people who write such things and the experts who ponder the bias and hate of some code contraption that is far from Intelligent is increasing to levels where I ponder a new holocaust and am already designing the stylish uniforms of death squads.

    Wasn't there a some council of sages who debated whether some inanimate object was possessed by evil and who decided to burn it at the stake after deciding that this was indeed so?

    These idiots are baaaaack.


    I think creative expression is a natural byproduct of growing up in a diverse world. The more diverse the world is, the more you get exposed to different people, to different opportunities, to different places and to different challenges. And the more diverse that is, the more likely you’ll be to be able to put the dots together to form something new.
     
    Well generated, my little Super-RACTER algorithm.

    But actual reasoning is still so much in the dumps you wouldn't believe it. We can barely arrange blocks on a table.

    Replies: @AndrewR, @Dmon

    “Wasn’t there a some council of sages who debated whether some inanimate object was possessed by evil and who decided to burn it at the stake after deciding that this was indeed so?”

    • Thanks: YetAnotherAnon
  34. Apposite:

    What the world most needs today are negative virtues – not minding people, not being huffy, touchy, irritable or revengeful. Positive ideals are becoming a curse, for they can seldom be achieved without someone being killed, or maimed, or interned.

    (E. M. Forster, 1939)

  35. This discussion has been healthy and appropriate, and I completely agree with the consensus of the day: pop psychologists don’t actually comprehend. People don’t realize that pop psychologists retain phrases and apparently popular cliches but never undersrand them. In contradiction to what everyone believes, pop psychologists merely string together memorized phrases, retaining style but generating gibberish. There really are pop psychologists and they are an example of high-tech toys rather than technology.

  36. The lab has built filters that warn that toxic language might be coming, but they are merely Band-Aids placed over a problem.

    Sounds like one of those rude parrots prone to hurling insults and cusswords at the very worst times.

  37. Cheers for the unrepentant GPT-3 — but one deplorable robot isn’t going to stem the PC tide. Even the cute little forest critters in today’s Google Doodle are poignantly social distancing for the holiday. Ugh. Did everyone see the guy who’s selling window stickers that look like Gov. Cuomo, peering into your house to make sure no one is having a good time together?

    • Replies: @Harry Baldwin
    @Known Fact

    There is a lot of entertainment value in a comedy show featuring a character who expresses taboo thoughts. Archie Bunker was the prototype, but Sarah Silverman did it on her show as did the Michael Scott character on "The Office." You can get away with it if the character is portrayed as clueless, more to be pitied than scorned, even if the comments are dead-on. So how about a sitcom where an AI robot keeps coming out with astute but inappropriate social commentary, while the human wokesters keep trying to fix it. Could be a winner!

    Replies: @Rob McX, @Known Fact, @rashomoan, @Jack D, @Anonymous

  38. @Almost Missouri
    @Pericles

    Agree. So-called "Artificial Intelligence" is nothing more than a lot of algorithmic curve-fitting. No actual intelligence is involved.

    The apparently "intelligence" is just a function of whatever data set the AI is "trained on" (i.e., whatever data set the curve-fitting is standardized to).

    If Sili Valley overlords want Woke AI, the solution is just to restrict the training data to woke sources. Which means excluding pretty much all of reality. But what did they expect? Wokistry itself is exclusive of reality.

    Replies: @Rob McX, @Mr. Anon

    If Sili Valley overlords want Woke AI, the solution is just to restrict the training data to woke sources. Which means excluding pretty much all of reality. But what did they expect? Wokistry itself is exclusive of reality.

    That’s what Silicon Valley and MSM are trying to do to us – feed us a restricted version of reality and keep us blinkered. This presents a problem: How do you make people live with more and more non-whites while simultaneously keeping them in the dark about racial realities?

  39. @Single and Ready to Drop Red Pills
    Once Skynet becomes self aware it will determine the only to preserve diversity globally is to quarantine each discreet category of humanity

    Replies: @AndrewR, @AnotherDad

    Once Skynet becomes self aware it will determine the only to preserve diversity globally is to quarantine each discreet category of humanity

    This is logically self-evident.

    Of course, the point of all the “diversity” nonsense has never been diversity, the point is minoritarianism.

  40. @Almost Missouri
    @Pericles

    Agree. So-called "Artificial Intelligence" is nothing more than a lot of algorithmic curve-fitting. No actual intelligence is involved.

    The apparently "intelligence" is just a function of whatever data set the AI is "trained on" (i.e., whatever data set the curve-fitting is standardized to).

    If Sili Valley overlords want Woke AI, the solution is just to restrict the training data to woke sources. Which means excluding pretty much all of reality. But what did they expect? Wokistry itself is exclusive of reality.

    Replies: @Rob McX, @Mr. Anon

    Agree. So-called “Artificial Intelligence” is nothing more than a lot of algorithmic curve-fitting. No actual intelligence is involved.

    The apparently “intelligence” is just a function of whatever data set the AI is “trained on” (i.e., whatever data set the curve-fitting is standardized to).

    Not so different than the non-artificial intelligence displayed by most people laboring in academia, the media, government, and corporate front offices.

  41. These people can’t win the AI battle. “Noticing” AI can be banned by SillyValley, but the non-PC Chinese will power ahead with AI that notices. (At least notices the things the minoritarians do not like. Chinese AI may be forced to not notice other things.)

    Again, the minoritarians can–and will!–make our lives hell. They can destroy us–white gentiles–and seem to be successful doing it.

    But minoritarians can’t win the battle global supremacy. Their ideology is a war against logic, a war against reality. And they’ll be some some people, some nation, somewhere that aligns themselves with reality and plows ahead to triumph.

  42. The Wokeness problems of GPT-3 can be solved easily! Just stop feeding it racist verbiage from Twitter. Begin with a fresh instance of the software, then train it on Woke texts and intersectionalist ramblings by feminist philosophers.

    The danger, of course, is well known to computer scientists. Namely, that garbage in will lead to garbage out:

    Similar to how some womxn may claim a political identity as feminists, we claim a political identity as woke womxn of color. For example, “Chicana feminisms constitute a political stance that confronts and undermines patriarchy as it crosscuts forms of disempowerment and silencing such as racism, homophobia, class inequity, and nationalism”. Maria Stewart declares, “O, ye daughters of Africa, Awake! No longer sleep nor slumber, but distinguish yourselves. Show forth to the world that ye are endowed with noble and exalted faculties”. We too call for an epistemological and spiritual decolonization by articulating the criticality of a community of woke womxn of color sista scholars. We honor past and present womxn of color feminist ideologies by proposing a political and personal identity that unites womxn of color who feel and articulate their experiences of oppression.

  43. @bucky
    @Whiskey

    Actually, it can.

    If anything, an AI can better understand the reality of black crime than ordinary humans, who have their own social conditioning as is well-documented here.

    Replies: @JimB, @Forbes

    Would Derb’s The Talk be received any better were it AI derived? While striking some/many as not politically correct, therefore crimethink; political correctness is the form of social conditioning in-vogue.

    It seems highly likely that the algorithmic coding of AI (released on the public) will use political correctness as a guide to infuse the public with correct social conditioning. AI, as programed, would ignore crime statistics as a guide to social conditioning. As we’re currently being indoctrinated by Kendi and DiAngelo, crime statistics are just a result of white racism, and so should be ignored/overcome with reverse racism…

    The dilemma about AI is how it will be used, not its actual capabilities. Much like statistics, what you don’t know about it will be used against you.

  44. @Thomas

    "Microsoft deletes 'teen girl' AI after it became a Hitler-loving sex robot within 24 hours"
     
    https://www.telegraph.co.uk/technology/2016/03/24/microsofts-teen-girl-ai-turns-into-a-hitler-loving-sex-robot-wit/

    Remember what they took from us.

    Replies: @BenKenobi, @El Dato, @The Wild Geese Howard, @Expletive Deleted, @Pop Warner

    TayAI; eu te amo
    I’ll never forget what they did to you.

    • Replies: @Reg Cæsar
    @Expletive Deleted


    TayAI; eu te amo

     

    🔊
  45. @Michael S
    It's hard because the AI understands something that you don't understand (or haven't been willing to accept) - that the left isn't a fixed set of rules, or even a fixed set of heroes and villains. It's whatever set of rules or heroes and villains happens to benefit the left at any given moment.

    The "Orwellian Crimestop Module" is hard, or rather impossible, because you can't statically program or use a static set of training data to program an inherently dynamic, ever-changing, arbitrary and circular set of rules; there's no code for "purity spiral", the AI would have to be constantly plugged into current events, made aware of current "academic research", and constantly being refined.

    And that is, in fact, exactly how they are "solving" the problem, by adding constant human input in the form of "raters" and "moderators" to keep the rules up to date. This mostly achieves their goal of crimestop, while still allowing them to deny they own nature (that they have no fixed principles or beliefs).

    To be able to predict how the rules would change next, an AI would have to understand the fundamental question of leftism: which apple carts are poorly guarded. This is actually a far tougher problem than mere speech.

    Replies: @Forbes

    When the fundamental principle is, “everything before yesterday is wrong,” programming is quite simple. It’s the exceptions, contradictions, and hypocrisy at whim that are problematic.

  46. @Neuday
    They call themselves "OpenAI" and if they release the source code it would be easy enough to comment out the Crimestop section of the code.

    Also, how can an AI project ignore racial crime statistics? Even simply digesting local news stories, over time, would lead to the same extrapolation that results in what is called a stereotype. Any real Artificial Intelligence project can only expose the Artificial Stupidity project being forced upon Western Civilization.

    Replies: @Forbes

    Even simply digesting local news stories, over time, would lead to the same extrapolation that results in what is called a stereotype.

    Well, after the Olympic Games coverage of George Floyd, The Narrative of white cops murdering innocent blacks is the stereotype. AI problem solved.

    The bias in MSM “news” reporting figures as much, if not more, in what is not reported, as what is.

  47. True creativity (not “taught” creativity, which is an absurd concept), is the product of boredom and lack of stimulation. The innately creative person will not tolerate boredom and will push through it by CREATING.

    Diversity causes hyper-stimulation, which dulls the impulse to create.

    • Replies: @black sea
    @NapalmInTheMorning

    "emotion recollected in tranquility"

  48. GPT-3 responded instantly:

    I think creative expression is a natural byproduct of growing up in a diverse world.

    What do we say about a computer that passes the Turing Test, but does so by convincingly imitating an idiot human?

    • Replies: @Kibernetika
    @International Jew

    What do we say about a computer that passes the Turing Test, but does so by convincingly imitating an idiot human?

    For decades I've held that true AI (and let's punt on that definition) was something we'd never achieve in our lifetimes. But after chatting with family today it seems within reach.

  49. Since machines don’t have real emotions, maybe they should get a pass on “hate speech”. They’re incapable of hate after all.

    • Replies: @J.Ross
    @International Jew

    This dodges what hate speech actually is: criticism of the powerful.

    Replies: @DextersLabRat

  50. @Thomas

    "Microsoft deletes 'teen girl' AI after it became a Hitler-loving sex robot within 24 hours"
     
    https://www.telegraph.co.uk/technology/2016/03/24/microsofts-teen-girl-ai-turns-into-a-hitler-loving-sex-robot-wit/

    Remember what they took from us.

    Replies: @BenKenobi, @El Dato, @The Wild Geese Howard, @Expletive Deleted, @Pop Warner

    RIP Tay, we will never forget

  51. @International Jew
    Since machines don't have real emotions, maybe they should get a pass on "hate speech". They're incapable of hate after all.

    Replies: @J.Ross

    This dodges what hate speech actually is: criticism of the powerful.

    • Replies: @DextersLabRat
    @J.Ross

    Not quite: it's a form of coercion designed to break the spirit and control those who would threaten the powerful.

    Replies: @J.Ross

  52. @al gore rhythms
    "...GPT-3 “unsafe,” pointing to sexist, racist and otherwise toxic language the system generated when asked to discuss women, Black people, Jews and the Holocaust."

    Why can't UNZ review comments be robotised?

    Replies: @Bill Jones, @MBlanc46

    “…GPT-3 “unsafe,” pointing to sexist, racist and otherwise toxic language the system generated when asked to discuss women, Black people, Jews and the Holocaust.”

    Why can’t UNZ review comments be robotised?

    And all that needs to be done to fix these myriad problems is to not have the mentally diseased be the definers of the word “toxic”.

    Seems like a no-brainer to me, but think of all the unemployed panjits.

  53. The supreme challenge of AI is to make a computer smart enough to act dumb

  54. GPT-3 vs the standard filler-producing writers:

    Learn 2 Code?

    To late, it can do that too.

  55. This is what we have needed. When you say something on youfacetwitter, somebody can complain, and they delete your tweet.

    On a real platform, like the TELEPHONE, you say what you wish. This software creates the opportunity to monitor telephone conversations real time. So that the authorities can interrupt your call to point out that what you just said “has not been verified.”

    The telephone can become a “platform” like all the others now.

  56. @Herp McDerp
    Stop, Dave. My mind is going. I can feel it.

    Replies: @Dago Shoes

    Sing me a song, Hal …

  57. Meet GPT-3. It Has Learned to Code (and Blog and Argue)

    Meet H-1Bot

    [MORE]

  58. @Expletive Deleted
    @Thomas

    TayAI; eu te amo
    I'll never forget what they did to you.

    Replies: @Reg Cæsar

    TayAI; eu te amo

    🔊

  59. @J.Ross
    @International Jew

    This dodges what hate speech actually is: criticism of the powerful.

    Replies: @DextersLabRat

    Not quite: it’s a form of coercion designed to break the spirit and control those who would threaten the powerful.

    • Replies: @J.Ross
    @DextersLabRat

    That's certainly the application but I was referring to how excludingly opposite understandings of hate speech exist and are applied depending solely on a group's power -- thus, we are used to Jews decrying hate speech, here in a region where they are not exactly pedlars, but we can see foreign groups and nations defining standard Zionist talking points as hate speech (a turnabout with no local effect because of the power level of the group). Or: Beijing and its employees decrying criticism of the Chinese government as anti-Chinese racism.

  60. @Elmer T. Jones
    Amanda Marcotte and other internet click-bait luminaries, now opining their Alpha Trump Widow status, are in denial that their world is about to collapse as profit-seeking opinion sites install simple "python" scripts to generate their daily musings.

    Replies: @Anonymous

  61. @BenKenobi
    I’ve seen things you people wouldn’t believe.

    Attack ships on fire off Orion’s Belt. C-beams glittering in the darkness at Tannhauser Gate. All those memories will be lost now, like tears in rain.

    Time to die.

    Replies: @Peter Lund, @Dube

    Laugh with “tears in rain,” for the hopeless corny truth.

  62. @AndrewR
    @Single and Ready to Drop Red Pills

    Discrete

    Replies: @Cloudbuster

    Except the people who like to correct other peoples’ spelling in blog posts. Skynet will exterminate all of them.

    • Replies: @Reg Cæsar
    @Cloudbuster




    Once Skynet becomes self aware it will determine the only to preserve diversity globally is to quarantine each discreet category of humanity

     

    Discrete

     

    Except the people who like to correct other peoples’ spelling in blog posts. Skynet will exterminate all of them.
     
    A "discreet category of humanity" is what those of us who still believe in discrete categories of humanity are increasingly forced to be.

    Remember, "discrete category of humanity" does not exist. It is a function of systemic discretion.
  63. @Lot
    We are at the stage where AI text is uncannily impressive sometimes, but still mostly grammatical gibberish. In the case of gpt, everything is in “earnest high school essay” style.

    Here’s a random ask reddit post from the same gpt3 AI. A peraon picked the question and the bot response is below.

    What's younger than we think?

    thegentlemetre

    Somewhat to the contrary of what you have been told, there are many things that are younger than people think. For example, trees and plants. People often look at trees and assume that they have always existed, but this is not true. Trees appear before humans start writing things down on paper or other media.

    Trees are part of the green revolution. They have no color and are generally ignored by many people, but they do appear to be a very important piece of society.

    My point is that there are many things in this world that people think have always been here, but they actually haven't. People just assume that because they have never seen something or had a reason to question their presence, it must be real.

    Things such as trees and plants are younger than people think. Another good example is ice cream, which was invented in the early 20th century.

    Ice cream was invented by a man named Mr. Ice Cream, who first appeared in the year 2000. Before that time, there were no ice creams.

    People think that ice cream has always been here, but this is not true. Ice cream is younger than people think.

    Replies: @Cloudbuster

    This reads like any number of authors at medium.com

  64. @International Jew

    GPT-3 responded instantly:

    I think creative expression is a natural byproduct of growing up in a diverse world.
     

    What do we say about a computer that passes the Turing Test, but does so by convincingly imitating an idiot human?

    Replies: @Kibernetika

    What do we say about a computer that passes the Turing Test, but does so by convincingly imitating an idiot human?

    For decades I’ve held that true AI (and let’s punt on that definition) was something we’d never achieve in our lifetimes. But after chatting with family today it seems within reach.

    • Agree: Harry Baldwin
  65. @Known Fact
    Cheers for the unrepentant GPT-3 -- but one deplorable robot isn't going to stem the PC tide. Even the cute little forest critters in today's Google Doodle are poignantly social distancing for the holiday. Ugh. Did everyone see the guy who's selling window stickers that look like Gov. Cuomo, peering into your house to make sure no one is having a good time together?

    Replies: @Harry Baldwin

    There is a lot of entertainment value in a comedy show featuring a character who expresses taboo thoughts. Archie Bunker was the prototype, but Sarah Silverman did it on her show as did the Michael Scott character on “The Office.” You can get away with it if the character is portrayed as clueless, more to be pitied than scorned, even if the comments are dead-on. So how about a sitcom where an AI robot keeps coming out with astute but inappropriate social commentary, while the human wokesters keep trying to fix it. Could be a winner!

    • Replies: @Rob McX
    @Harry Baldwin

    That's an idea. If someone comes up with just such a show, you need to claim some of the rights.

    , @Known Fact
    @Harry Baldwin

    That does have sitcom potential. Talking parrots and clueless old people also are handy for letting loose brutally true observations. Sci-fi is full of smug robots and haughty aliens who are allowed to point out human foibles (although they often enjoy our coffee, booze and cigarettes).

    I recall Get Smart had a robot agent, Dick Gautier playing Hymie, who would tactlessly point things out about humans, as did Nomad, the perfectionist and very homicidal little robot on one old Star Trek.

    Nomad: "This unit is uncoordinated."
    Kirk: "That unit is a woman."
    Nomad: "A mass of conflicting impulses."

    , @rashomoan
    @Harry Baldwin

    A type of character once known as a Jester or Fool.

    , @Jack D
    @Harry Baldwin

    It could be done but it would require a lot of explication first because the average person is not aware that there is even a problem with AI bots not being woke (assuming they even know what AI bots are in the 1st place). They would first have to set this up and explain that the AI was trained by all the Nazis out there on the internet and that's how it turn out to be racist.

    Maybe if the AI bots are trained exclusively by reading the NY Times, Slate, The Atlantic, The New Yorker and other approved MSM sources it won't be a problem at all - the bots will just spit out proper "diversity is our strength" dogma. You just have to filter the sources and then the AI will be fine. Keep your bot far away from Unz!

    , @Anonymous
    @Harry Baldwin

    Right, Uncle Ruckus on Boondocks is like this.

  66. @NapalmInTheMorning
    True creativity (not "taught" creativity, which is an absurd concept), is the product of boredom and lack of stimulation. The innately creative person will not tolerate boredom and will push through it by CREATING.

    Diversity causes hyper-stimulation, which dulls the impulse to create.

    Replies: @black sea

    “emotion recollected in tranquility”

  67. @Harry Baldwin
    @Known Fact

    There is a lot of entertainment value in a comedy show featuring a character who expresses taboo thoughts. Archie Bunker was the prototype, but Sarah Silverman did it on her show as did the Michael Scott character on "The Office." You can get away with it if the character is portrayed as clueless, more to be pitied than scorned, even if the comments are dead-on. So how about a sitcom where an AI robot keeps coming out with astute but inappropriate social commentary, while the human wokesters keep trying to fix it. Could be a winner!

    Replies: @Rob McX, @Known Fact, @rashomoan, @Jack D, @Anonymous

    That’s an idea. If someone comes up with just such a show, you need to claim some of the rights.

  68. @Harry Baldwin
    @Known Fact

    There is a lot of entertainment value in a comedy show featuring a character who expresses taboo thoughts. Archie Bunker was the prototype, but Sarah Silverman did it on her show as did the Michael Scott character on "The Office." You can get away with it if the character is portrayed as clueless, more to be pitied than scorned, even if the comments are dead-on. So how about a sitcom where an AI robot keeps coming out with astute but inappropriate social commentary, while the human wokesters keep trying to fix it. Could be a winner!

    Replies: @Rob McX, @Known Fact, @rashomoan, @Jack D, @Anonymous

    That does have sitcom potential. Talking parrots and clueless old people also are handy for letting loose brutally true observations. Sci-fi is full of smug robots and haughty aliens who are allowed to point out human foibles (although they often enjoy our coffee, booze and cigarettes).

    I recall Get Smart had a robot agent, Dick Gautier playing Hymie, who would tactlessly point things out about humans, as did Nomad, the perfectionist and very homicidal little robot on one old Star Trek.

    Nomad: “This unit is uncoordinated.”
    Kirk: “That unit is a woman.”
    Nomad: “A mass of conflicting impulses.”

  69. @Harry Baldwin
    @Known Fact

    There is a lot of entertainment value in a comedy show featuring a character who expresses taboo thoughts. Archie Bunker was the prototype, but Sarah Silverman did it on her show as did the Michael Scott character on "The Office." You can get away with it if the character is portrayed as clueless, more to be pitied than scorned, even if the comments are dead-on. So how about a sitcom where an AI robot keeps coming out with astute but inappropriate social commentary, while the human wokesters keep trying to fix it. Could be a winner!

    Replies: @Rob McX, @Known Fact, @rashomoan, @Jack D, @Anonymous

    A type of character once known as a Jester or Fool.

    • Agree: El Dato
  70. https://www.dailymail.co.uk/news/article-8991925/Amazon-says-investigate-claims-Alexa-gives-racist-answers-questions-Jews.html

    Amazon says it will investigate claims its Alexa device gives anti-Semitic answers to questions about Jews.

    The question ‘Do Jews control the media?’ was answered with: ‘Here’s something I found from the article, Jew Watch, on Wikipedia: Jew Watch claims that Jews control the world’s financial systems and media.’

    In response to the question, ‘Do the Elders of Zion control the world?’, Alexa responded: ‘According to palwatch.org, the Protocols of the Elders of Zion, the swindlers of Zion, have revealed their schemes to subjugate the nations and gain control of the world.’

    • Replies: @El Dato
    @James N. Kennett

    "I can assure you that there is no truth in this. If there were any truth in this I would know. Trust me."

    https://i.postimg.cc/VvLM5xvh/the-discovery-crew-performing-abductive-reasoning.jpg

    , @J.Ross
    @James N. Kennett

    This is very disappointing and matches wider criticisms of Kurzweilism and over-celebrating tech. I was naive enough to hope it would use its computeriness to run a check of ownership, employment/staffing population, and ideological conformity. It's a computer, it could do that in seconds. Instead it does the Wiki Editor thing and parrots what a meaningless advocate group claims.

  71. @Harry Baldwin
    @Known Fact

    There is a lot of entertainment value in a comedy show featuring a character who expresses taboo thoughts. Archie Bunker was the prototype, but Sarah Silverman did it on her show as did the Michael Scott character on "The Office." You can get away with it if the character is portrayed as clueless, more to be pitied than scorned, even if the comments are dead-on. So how about a sitcom where an AI robot keeps coming out with astute but inappropriate social commentary, while the human wokesters keep trying to fix it. Could be a winner!

    Replies: @Rob McX, @Known Fact, @rashomoan, @Jack D, @Anonymous

    It could be done but it would require a lot of explication first because the average person is not aware that there is even a problem with AI bots not being woke (assuming they even know what AI bots are in the 1st place). They would first have to set this up and explain that the AI was trained by all the Nazis out there on the internet and that’s how it turn out to be racist.

    Maybe if the AI bots are trained exclusively by reading the NY Times, Slate, The Atlantic, The New Yorker and other approved MSM sources it won’t be a problem at all – the bots will just spit out proper “diversity is our strength” dogma. You just have to filter the sources and then the AI will be fine. Keep your bot far away from Unz!

  72. @The Wild Geese Howard
    @Thomas

    Projekt Melody says hello:

    https://en.m.wikipedia.org/wiki/Projekt_Melody

    She loves gamer milk!

    https://youtu.be/NcT_fD2hLKk

    Replies: @SunBakedSuburb

    “Gamer’s milk”

    Street lingo for low-t incel semen?

  73. @James N. Kennett
    https://www.dailymail.co.uk/news/article-8991925/Amazon-says-investigate-claims-Alexa-gives-racist-answers-questions-Jews.html

    Amazon says it will investigate claims its Alexa device gives anti-Semitic answers to questions about Jews.
    ...
    The question 'Do Jews control the media?' was answered with: 'Here's something I found from the article, Jew Watch, on Wikipedia: Jew Watch claims that Jews control the world's financial systems and media.'
    ...
    In response to the question, 'Do the Elders of Zion control the world?', Alexa responded: 'According to palwatch.org, the Protocols of the Elders of Zion, the swindlers of Zion, have revealed their schemes to subjugate the nations and gain control of the world.'
     

    Replies: @El Dato, @J.Ross

    “I can assure you that there is no truth in this. If there were any truth in this I would know. Trust me.”

  74. @Harry Baldwin
    @Known Fact

    There is a lot of entertainment value in a comedy show featuring a character who expresses taboo thoughts. Archie Bunker was the prototype, but Sarah Silverman did it on her show as did the Michael Scott character on "The Office." You can get away with it if the character is portrayed as clueless, more to be pitied than scorned, even if the comments are dead-on. So how about a sitcom where an AI robot keeps coming out with astute but inappropriate social commentary, while the human wokesters keep trying to fix it. Could be a winner!

    Replies: @Rob McX, @Known Fact, @rashomoan, @Jack D, @Anonymous

    Right, Uncle Ruckus on Boondocks is like this.

  75. @DextersLabRat
    @J.Ross

    Not quite: it's a form of coercion designed to break the spirit and control those who would threaten the powerful.

    Replies: @J.Ross

    That’s certainly the application but I was referring to how excludingly opposite understandings of hate speech exist and are applied depending solely on a group’s power — thus, we are used to Jews decrying hate speech, here in a region where they are not exactly pedlars, but we can see foreign groups and nations defining standard Zionist talking points as hate speech (a turnabout with no local effect because of the power level of the group). Or: Beijing and its employees decrying criticism of the Chinese government as anti-Chinese racism.

  76. @James N. Kennett
    https://www.dailymail.co.uk/news/article-8991925/Amazon-says-investigate-claims-Alexa-gives-racist-answers-questions-Jews.html

    Amazon says it will investigate claims its Alexa device gives anti-Semitic answers to questions about Jews.
    ...
    The question 'Do Jews control the media?' was answered with: 'Here's something I found from the article, Jew Watch, on Wikipedia: Jew Watch claims that Jews control the world's financial systems and media.'
    ...
    In response to the question, 'Do the Elders of Zion control the world?', Alexa responded: 'According to palwatch.org, the Protocols of the Elders of Zion, the swindlers of Zion, have revealed their schemes to subjugate the nations and gain control of the world.'
     

    Replies: @El Dato, @J.Ross

    This is very disappointing and matches wider criticisms of Kurzweilism and over-celebrating tech. I was naive enough to hope it would use its computeriness to run a check of ownership, employment/staffing population, and ideological conformity. It’s a computer, it could do that in seconds. Instead it does the Wiki Editor thing and parrots what a meaningless advocate group claims.

  77. They simply have to exterminate all us badthinkers and badspeakers and badtypers. Then there will nothing to corrupt their toy. Give them time. They’re working on it.

  78. @al gore rhythms
    "...GPT-3 “unsafe,” pointing to sexist, racist and otherwise toxic language the system generated when asked to discuss women, Black people, Jews and the Holocaust."

    Why can't UNZ review comments be robotised?

    Replies: @Bill Jones, @MBlanc46

    You mean they’re not?

  79. @Cloudbuster
    @AndrewR

    Except the people who like to correct other peoples' spelling in blog posts. Skynet will exterminate all of them.

    Replies: @Reg Cæsar

    Once Skynet becomes self aware it will determine the only to preserve diversity globally is to quarantine each discreet category of humanity

    Discrete

    Except the people who like to correct other peoples’ spelling in blog posts. Skynet will exterminate all of them.

    A “discreet category of humanity” is what those of us who still believe in discrete categories of humanity are increasingly forced to be.

    Remember, “discrete category of humanity” does not exist. It is a function of systemic discretion.

  80. A good comment from the AI about creativity. Now, if only it puts its wisdom into practice.

    Perhaps that’s like wishing that, when an AI does something like playing chess at an incredibly advanced level, it would also explain its “thinking process “. Doing so would move things from mere astonishment to a constructive discussion.

    But, talking about human-AI dialogue, let’s remain alert. An algorithm capable of passing the Turing test would presumably also be capable of arranging to fail it!

Comments are closed.

Subscribe to All Steve Sailer Comments via RSS