Jump to content

Recommended Posts

Posted
Just now, bolverk said:

Icono's wet dream comes true:

Monarchy at the top; then the Houses of Musk, Thiel, Zuckerberg, Bezos; followed by some sort of merchant class; and of course serfs and/or slaves at the bottom.

"Maybe" - they said the same thing about the internet.  How it was going to replace tens of millions of jobs, only to see hundreds of millions of new opportunities arise.  One of the best quotes I've heard about innovation and disruption was along the line of asking people a hundred+ years ago what they all wanted:  a faster horse.  

Posted
2 minutes ago, bolverk said:

Icono's wet dream comes true:

Monarchy at the top; then the Houses of Musk, Thiel, Zuckerberg, Bezos; followed by some sort of merchant class; and of course serfs and/or slaves at the bottom.

I mean, you won't even have the merchants or serfs/slaves.  AI is going to replace them all.  These guys are working towards a dream of 1) a future human society that 2) doesn't need or want humans.

Posted
1 minute ago, Brisketexan said:

I mean, you won't even have the merchants or serfs/slaves.  AI is going to replace them all.  These guys are working towards a dream of 1) a future human society that 2) doesn't need or want humans.

Mad Max shit, I guess, for the rest of us then.

Posted
4 minutes ago, BabaYaga said:

Hollywood is going to be even more bifurcated with their movie types - you'll have your lower budget shows that feature actors together focusing on plot and character development, and then all your big, blockbuster stuff will be mainly AI driven.  All your fantasy, superhero, Jurassic park, etc. stuff will be AI driven where plot and storytelling give way to animation and special effects.  

Don’t know if that’s your intent, but it’s interesting you treat animated film as “lesser than” art form.

4 minutes ago, Brisketexan said:

I mean, you won't even have the merchants or serfs/slaves.  AI is going to replace them all.  These guys are working towards a dream of 1) a future human society that 2) doesn't need or want humans.

The optimist in me hope for more of that Star Trek economy. 

Posted
2 minutes ago, YChang said:

Don’t know if that’s your intent, but it’s interesting you treat animated film as “lesser than” art form.

The optimist in me hope for more of that Star Trek economy. 

Nope.

4 minutes ago, bolverk said:

Mad Max shit, I guess, for the rest of us then.

Yep.

You'll get the worst, because the people driving this are the worst.  They are genuine, unabashed and unfiltered sociopaths.  Not only do they have zero empathy or humanity, they actively decry empathy and humanity as flaws that must be stamped out.

We are being governed and controlled by the worst humans to ever live, who have the actual power and authority to end our species.  And they have proudly and repeatedly announced that's what they're using their power and authority to accomplish.  And we stand by and cheer them on as they work towards our extinction.  Dumbest fucking timeline.  But I guess on the bright side, it looks like it's going to be the FINAL timeline.  So, we've got that going for us, which is nice.

  • Hook 'Em 1
Posted
4 minutes ago, Brisketexan said:

Apparently, we're all-in on a plan of "the beast will consume everything, and then begin consuming itself, complaining about the fact that it's hungry now that there's nothing left to eat."  Fucking brilliant.

I gotta admit, I'm truly amazed at how rapidly we've accelerated to across-the-board goals of "let's destroy society and humanity completely" all in the service of -- for a little while, at least -- funneling all remaining wealth to a few hundred oligarchs.  The short-sightedness is something to behold.

I just feel really, really, really fucking bad for my kids.

I hear you.  An attempt is certainly being made, but that does not meant it will succeed.  Fight against the provision in the big beautiful bill that aims to bar states from regulating of AI for the next 10 years.

https://www.businessinsider.com/trump-big-beautiful-bill-ai-regulation-ban-tech-2025-6

Posted
3 minutes ago, YChang said:

Don’t know if that’s your intent, but it’s interesting you treat animated film as “lesser than” art form.

7 minutes ago, Brisketexan said:

Let's be fair, "most" blockbuster movies are light on plot & character development and high on explosions and actors dripping with HGH and peptides.  

Harry Potter doesn't fit this mold, but that is because it's a book adaptation.  Same with GOT.  

Posted
12 minutes ago, BabaYaga said:

Hollywood is going to be even more bifurcated with their movie types - you'll have your lower budget shows that feature actors together focusing on plot and character development, and then all your big, blockbuster stuff will be mainly AI driven.  All your fantasy, superhero, Jurassic park, etc. stuff will be AI driven where plot and storytelling give way to animation and special effects.  

As an indy musician, I've been using AI a lot for cover art and I am experimenting with AI generated videos.  The biggest issue I see right now is that it's hard to have a central or recurring character across images or scenes with video.  The stuff that looks great now are mostly collections of scenes of random humans doing or saying something.  It's difficult (maybe impossible given current state of AI?) to have something like a singular protagonist or villain that is present throughout a bunch of different scenes in a more comprehensive work.   When that can be overcome, that is when you'll wind see movie making disrupted.

Posted
6 minutes ago, Goredho said:

I hear you.  An attempt is certainly being made, but that does not meant it will succeed.  Fight against the provision in the big beautiful bill that aims to bar states from regulating of AI for the next 10 years.

https://www.businessinsider.com/trump-big-beautiful-bill-ai-regulation-ban-tech-2025-6

That's part of the unstoppable dismal tide to which I was referring.  We're hell-bent on social and species suicide, and to show our commitment to that goal, we'll ban all suicide-prevention activities for the next decade.

It's fucking insanity.  Collective insanity, towards a suicidal goal, which we are actively fucking cheering on.  What the fucking fuck is fucking wrong with us?

Posted
1 minute ago, Brisketexan said:

That's part of the unstoppable dismal tide to which I was referring.  We're hell-bent on social and species suicide, and to show our commitment to that goal, we'll ban all suicide-prevention activities for the next decade.

It's fucking insanity.  Collective insanity, towards a suicidal goal, which we are actively fucking cheering on.  What the fucking fuck is fucking wrong with us?

We're only human, after all.

Posted (edited)
37 minutes ago, BabaYaga said:

Hollywood is going to be even more bifurcated with their movie types - you'll have your lower budget shows that feature actors together focusing on plot and character development, and then all your big, blockbuster stuff will be mainly AI driven.  All your fantasy, superhero, Jurassic park, etc. stuff will be AI driven where plot and storytelling give way to animation and special effects.  

Quoting you again, as one of the experiments I tried with AI was to see how it might work at creating a graphic novel from Cormac McCarthy's "Blood Meridian".  I gave it this prompt:

Generate a graphic novel scene from this description of a commanche war band:

A legion of horribles, hundreds in number, half naked or clad in costumes attic or biblical or wardrobed out of a fevered dream with the skins of animals and silk finery and pieces of uniform still tracked with the blood of prior owners, coats of slain dragoons, frogged and braided cavalry jackets, one in a stovepipe hat and one with an umbrella and one in white stockings and a bloodstained wedding veil and some in headgear or cranefeathers or rawhide helmets that bore the horns of bull or buffalo and one in a pigeontailed coat worn backwards and otherwise naked and one in the armor of a Spanish conquistador, the breastplate and pauldrons deeply dented with old blows of mace or sabre done in another country by men whose very bones were dust and many with their braids spliced up with the hair of other beasts until they trailed upon the ground and their horses' ears and tails worked with bits of brightly colored cloth and one whose horse's whole head was painted crimson red and all the horsemen's faces gaudy and grotesque with daubings like a company of mounted clowns, death hilarious, all howling in a barbarous tongue and riding down upon them like a horde from a hell more horrible yet than the brimstone land of Christian reckoning, screeching and yammering and clothed in smoke like those vaporous beings in regions beyond right knowing where the eye wanders and the lip jerks and drools.

This is what it spit out:
image.thumb.png.7e9db74b4ad4b908ec29fa7c39522106.png

But, if I told it to generate two separate images of The Judge based on two different paragraphs from the book, the Judge would not be consistent across the images, even if I took the first image and used it to guide the generation of the image from the 2nd scene.  But once that is improved to the point of indistinguishability, you'll see AI entertainment take off.  There is a whole lot of creative work that is either un-copyrighted or in the public domain that could be used as prompt sources for AI generated entertainment.

Edited by Goredho
  • Hook 'Em 1
Posted
1 hour ago, Brisketexan said:

Apparently, we're all-in on a plan of "the beast will consume everything, and then begin consuming itself, complaining about the fact that it's hungry now that there's nothing left to eat."  Fucking brilliant.

I gotta admit, I'm truly amazed at how rapidly we've accelerated to across-the-board goals of "let's destroy society and humanity completely" all in the service of -- for a little while, at least -- funneling all remaining wealth to a few hundred oligarchs.  The short-sightedness is something to behold.

I just feel really, really, really fucking bad for my kids.

I get why people are AI doomers, but I think this is extremely short-sighted, myopic and even a bit silly.

Technological paradigm shifts are scary when you are 58 years old like we are. But the world is not ending and things will shake out and new technology and business norms will eventually become...normal and boring. Like you using a macbook and lexus nexus and doing discovery via e-filing or whatever in your day to day life versus paper and pen.

You sound a lot like the draftsman who bemoaned how 3 floors of hundreds of manual drafting drawers with a great degree and skill and healthy income were thrown out for one guy to do CAD.

Or the thousands of bookkeepers and manual recorders organizations employed who with an upper middle class life were eliminated overnight with the advents of ERPs.

Life will go on and new avenues will open. The only thing that makes this unique is the speed at which the transition is happening because AI seemingly popped overnight in 2022 and disruption is happening across the board at the same time. 

  • Hook 'Em 1
Posted
1 hour ago, Goredho said:

Quoting you again, as one of the experiments I tried with AI was to see how it might work at creating a graphic novel from Cormac McCarthy's "Blood Meridian".  I gave it this prompt:

Generate a graphic novel scene from this description of a commanche war band:

A legion of horribles, hundreds in number, half naked or clad in costumes attic or biblical or wardrobed out of a fevered dream with the skins of animals and silk finery and pieces of uniform still tracked with the blood of prior owners, coats of slain dragoons, frogged and braided cavalry jackets, one in a stovepipe hat and one with an umbrella and one in white stockings and a bloodstained wedding veil and some in headgear or cranefeathers or rawhide helmets that bore the horns of bull or buffalo and one in a pigeontailed coat worn backwards and otherwise naked and one in the armor of a Spanish conquistador, the breastplate and pauldrons deeply dented with old blows of mace or sabre done in another country by men whose very bones were dust and many with their braids spliced up with the hair of other beasts until they trailed upon the ground and their horses' ears and tails worked with bits of brightly colored cloth and one whose horse's whole head was painted crimson red and all the horsemen's faces gaudy and grotesque with daubings like a company of mounted clowns, death hilarious, all howling in a barbarous tongue and riding down upon them like a horde from a hell more horrible yet than the brimstone land of Christian reckoning, screeching and yammering and clothed in smoke like those vaporous beings in regions beyond right knowing where the eye wanders and the lip jerks and drools.

This is what it spit out:
image.thumb.png.7e9db74b4ad4b908ec29fa7c39522106.png

But, if I told it to generate two separate images of The Judge based on two different paragraphs from the book, the Judge would not be consistent across the images, even if I took the first image and used it to guide the generation of the image from the 2nd scene.  But once that is improved to the point of indistinguishability, you'll see AI entertainment take off.  There is a whole lot of creative work that is either un-copyrighted or in the public domain that could be used as prompt sources for AI generated entertainment.

You should really use a more fit-for-purpose AI tool than a general LLM like ChatGPT. Use MidJourney and narrow in on the prompting and you'd get something super fantastic is my guess.

  • Hook 'Em 1
Posted
8 minutes ago, Vegas64 said:

I get why people are AI doomers, but I think this is extremely short-sighted, myopic and even a bit silly.

Technological paradigm shifts are scary when you are 58 years old like we are. But the world is not ending and things will shake out and new technology and business norms will eventually become...normal and boring. Like you using a macbook and lexus nexus and doing discovery via e-filing or whatever in your day to day life versus paper and pen.

You sound a lot like the draftsman who bemoaned how 3 floors of hundreds of manual drafting drawers with a great degree and skill and healthy income were thrown out for one guy to do CAD.

Or the thousands of bookkeepers and manual recorders organizations employed who with an upper middle class life were eliminated overnight with the advents of ERPs.

Life will go on and new avenues will open. The only thing that makes this unique is the speed at which the transition is happening because AI seemingly popped overnight in 2022 and disruption is happening across the board at the same time. 

Or....and follow me here....I'm taking these people EXACTLY at their word:

Quote

Mechanize was launched on Thursday via a post on X by its founder, famed AI researcher Tamay Besiroglu. The startup’s goal, Besiroglu wrote, is “the full automation of all work” and “the full automation of the economy.” 

Does that mean Mechanize is working to replace every human worker with an AI agent bot? Essentially, yes. The startup wants to provide the data, evaluations, and digital environments to make worker automation of any job possible.

Besiroglu even calculated Mechanize’s total addressable market by aggregating all the wages humans are currently paid. “The market potential here is absurdly large: workers in the US are paid around $18 trillion per year in aggregate. For the entire world, the number is over three times greater, around $60 trillion per year,” he wrote.

Do they want to replace all human labor and eliminate all human wages?  They say they do.  We should fucking believe them.  We ignore their stated goals at our peril.

  • Hook 'Em 2
  • Like 1
Posted
2 minutes ago, Brisketexan said:

Or....and follow me here....I'm taking these people EXACTLY at their word:

Do they want to replace all human labor and eliminate all human wages?  They say they do.  We should fucking believe them.  We ignore their stated goals at our peril.

Do you believe, to the detriment of demonstrable reality and facts and history, every provocative thing you hear?

In just the recent past we have had customers claim they are going to get rid of people and use AI and completely walk that back.

Quote

Klarna, the Swedish fintech firm, initially replaced much of its customer service staff with AI chatbots, claiming they performed the work of 700 employees. However, due to declining service quality and customer dissatisfaction, the company is now rehiring human agents. CEO Sebastian Siemiatkowski acknowledged that overemphasis on cost-cutting led to poorer service and emphasized the necessity of human interaction for customer satisfaction. The new hiring strategy involves remote, contract-based roles targeting students and rural workers, raising concerns about job stability. This shift highlights the limitations of AI in handling complex, empathetic customer interactions and underscores the enduring value of human support (Source: Gizmodo)

The obvious reality is that yes, of course there will be some human labor diminshed and savings involved with using AI. Story as old as time.

But with new business and tech shifts, new business models and human working models will arise. 

 

Posted
11 minutes ago, Vegas64 said:

I get why people are AI doomers, but I think this is extremely short-sighted, myopic and even a bit silly.

Technological paradigm shifts are scary when you are 58 years old like we are. But the world is not ending and things will shake out and new technology and business norms will eventually become...normal and boring. Like you using a macbook and lexus nexus and doing discovery via e-filing or whatever in your day to day life versus paper and pen.

You sound a lot like the draftsman who bemoaned how 3 floors of hundreds of manual drafting drawers with a great degree and skill and healthy income were thrown out for one guy to do CAD.

Or the thousands of bookkeepers and manual recorders organizations employed who with an upper middle class life were eliminated overnight with the advents of ERPs.

Life will go on and new avenues will open. The only thing that makes this unique is the speed at which the transition is happening because AI seemingly popped overnight in 2022 and disruption is happening across the board at the same time. 

I mean, did you read the article Brisket was responding to?  It's an AI company whose mission statement is literally to automate away all human labor.  Not an industry.  All industries.  Starting with white collar jobs, and then manual labor when robotics catches up.  And their launch is coming at the same time that  there is a push for cementing zero regulation of any AI efforts for a decade.  The new avenues opening up you posit?  That's just another human labor target for them to automate away.

That's not hyperbole.  It's their stated intent.  Will they succeed?  I certainly hope not.

Posted
Just now, Vegas64 said:

Do you believe, to the detriment of demonstrable reality and facts and history, every provocative thing you hear?

Do I believe it, as in believe with 100% certainty it will happen?  Of course not.

When it is a matter of sufficient severity, do I treat it as a threat worthy of our attention?  Hell yes.

If the representation is "man, Brisket, if I ever met you in a bar, I'd probably punch your face in," I'm not terribly alarmed (shit, I've met me -- I probably couldn't blame you).  If the representation is "Brisket, I am coming to your house tonight, at [my actual address], and I'm going to kill you and your wife as you sleep".....yeah, I'm legit alarmed.  The severity of the outcome at risk absolutely figures in to how seriously we should take it.

"Hey, I intend to eliminate all human life, and here's how I'm going to do it," said by someone who has even an outside chance of accomplishing it?  Fucking pay attention to that shit.

Jesus tapdancing Christ, we've spent the last 10+ years having person after person tell us we're being alarmist when there are threats to do things like, oh, I don't know, end birthright citizenship, repeal Roe v. Wade, use the DOJ as a personal law firm to pursue one man's vendettas, etc. etc. etc.....only to be fucking right about all of that shit.  So, fuck your "you're just being alarmist" shit when there are billionaire oligarchs saying "Brisket, I am going to make sure your kids are unemployed, and unemployable, and have no means to make their way in this world."  That's a clear and direct threat against my family and all of humanity.  Take it fucking seriously.

  • Hook 'Em 1
Posted
5 minutes ago, Goredho said:

I mean, did you read the article Brisket was responding to?  It's an AI company whose mission statement is literally to automate away all human labor.  Not an industry.  All industries.  Starting with white collar jobs, and then manual labor when robotics catches up.  And their launch is coming at the same time that  there is a push for cementing zero regulation of any AI efforts for a decade.  The new avenues opening up you posit?  That's just another human labor target for them to automate away.

That's not hyperbole.  It's their stated intent.  Will they succeed?  I certainly hope not.

It's one voice. And if it becomes the reality then it goes against everything we've ever seen or known in the history of technology in the world. Statistically so improbable it's completely irrational to use that as a buttress for your belief and worldview IMO.

The story of technology going back into forever (since literally dawn of man) is that OF COURSE technology displaces jobs, but the history of all of humanity is we find new things to do pretty effectively. Even Brisket. And especially Brisket's children and their children.

Posted
1 minute ago, Vegas64 said:

It's one voice. And if it becomes the reality then it goes against everything we've ever seen or known in the history of technology in the world. Statistically so improbable it's completely irrational to use that as a buttress for your belief and worldview IMO.

The story of technology going back into forever (since literally dawn of man) is that OF COURSE technology displaces jobs, but the history of all of humanity is we find new things to do pretty effectively. Even Brisket. And especially Brisket's children and their children.

Well, certainly the smart thing to do is....completely ban all regulation of the technology for the next ten years.  Yep, nothing to see here, no worries at all.  Just a technology that more than one fucking person (he ain't the only one) says they intend to use to completely end human society and economies, with no replacement in mind.....and a government so in thrall to and captured by those voices that it is going to ban all regulation of that technology.

Goddamn, you'd chug the fucking kool-aid every damned time, wouldn't you?

Posted
13 minutes ago, Vegas64 said:

It's one voice. And if it becomes the reality then it goes against everything we've ever seen or known in the history of technology in the world. Statistically so improbable it's completely irrational to use that as a buttress for your belief and worldview IMO.

The story of technology going back into forever (since literally dawn of man) is that OF COURSE technology displaces jobs, but the history of all of humanity is we find new things to do pretty effectively. Even Brisket. And especially Brisket's children and their children.

Do you mind sharing what you do for a living?  I'm curious.  I'm a software/devops and data engineer.

Posted
7 minutes ago, Brisketexan said:

Well, certainly the smart thing to do is....completely ban all regulation of the technology for the next ten years.  Yep, nothing to see here, no worries at all.  Just a technology that more than one fucking person (he ain't the only one) says they intend to use to completely end human society and economies, with no replacement in mind.....and a government so in thrall to and captured by those voices that it is going to ban all regulation of that technology.

Goddamn, you'd chug the fucking kool-aid every damned time, wouldn't you?

I'm actually not sure where I stand on the regulation stuff yet. The bill as I understand it is banning a patchwork of regulations via state laws. And even then states could enforce their own AI regulation, they just would be locked out of some AI funding and otehr stuff, right?

Posted
3 minutes ago, Chult86 said:

At this point if you have any white collar positions at your company and you're not using ChatGPT or some form of Generative text, you're falling behind.

ChatGPT is the bare minimum. For whatever role you have there are now fit-for-purpose AI tools to help up. There is a funny meme out there of the YC slides "Cursor for [enter your department/vertical]"

Posted (edited)
7 minutes ago, Vegas64 said:

I'm actually not sure where I stand on the regulation stuff yet. The bill as I understand it is banning a patchwork of regulations via state laws. And even then states could enforce their own AI regulation, they just would be locked out of some AI funding and otehr stuff, right?

Correct in that it is incremental.  Step 1: ban all state regulation.  Step 2: full capture of Congress and the federal government so ensure no meaningful federal regulation of AI.

Dude.  Step 2 is already accomplished.  Step 1 is actually second in the order of operations, thus ensuring the overarching goal: no meaningful regulation of AI.

We are going to do more to make sure that no teenager can see tits on the internet than we're going to do to prevent AI and the internet from ensuring that teenager never has a job.

Edited by Brisketexan
Posted
1 minute ago, Brisketexan said:

Correct in that it is incremental.  Step 1: ban all state regulation.  Step 2: full capture of Congress and the federal government so ensure no meaningful federal regulation of AI.

Dude.  Step 2 is already accomplished.  Step 1 is actually second in the order of operations, thus ensuring the overarching goal: no meaningful regulation of AI.

We are going to do more to make sure that no teenager can see tits on the internet than we're going to do to prevent AI and the internet from ensuring that teenager never has a job.

What do you mean a teenager never has a job? Do you think that AI is going to render teenager jobs and summer jobs useless?

The reason I think relaxing regulation on AI in the storming, norming and forming phase of AI going from hype cycle to mature is because that is how technology, historically, gets to an optimal state of value. The invisible hand of capitalism will have influence as well.

AI and tech supremacy can be argued as a zero sum game. And if we (over)regulate today, there is real risk of losing the AI race to countries who won't (e.g. China, India, EU, Israel, etc.). To deny regulation hampers competitiveness is to deny the very "just the facts ma'am" type reality and pragmatism you claim to be serving in your AI doom.

Posted
17 minutes ago, Goredho said:

Do you mind sharing what you do for a living?  I'm curious.  I'm a software/devops and data engineer.

I'm a retired history teacher who has too much time on his hands and super interested in technology from an angel investor perspective and from  a sociological perspective (just bought Daniel Susskind's new book, A World Without Work).

As someone who is not technical, I've found AI to be very liberating. It's not quite (and never will be) something that levels the playing field with someone like you who is very technical, but the number of semantic agents, vibe coding, and AI-enabling tools is very democratizing. 

There are very few real transformational periods to live through. PC revolution and mobile were obviously in our lifetime. AI is happening now and it's messy and painful but also beautiful and glorious.

  • Hook 'Em 3
Posted
27 minutes ago, Vegas64 said:

As someone who is not technical, I've found AI to be very liberating. It's not quite (and never will be) something that levels the playing field with someone like you who is very technical, but the number of semantic agents, vibe coding, and AI-enabling tools is very democratizing. 

There are very few real transformational periods to live through. PC revolution and mobile were obviously in our lifetime. AI is happening now and it's messy and painful but also beautiful and glorious.

It's pretty remarkable what you can accomplish now with an agent to help with implementation and limited troubleshooting. With that said, it's also going to lead to a massive explosion in orphaned and unexplainable code within organizations. I'm currently, right now, dealing with that problem with a customer - only they got their unmaintainable code thanks to south asian contracting firms. Most of the jobs that genAI is going to replace first is that tier of contracting job, since the work product is equivalent and there's much less needful that needs doing.

  • Hook 'Em 1
Posted
6 minutes ago, Captainant said:

It's pretty remarkable what you can accomplish now with an agent to help with implementation and limited troubleshooting. With that said, it's also going to lead to a massive explosion in orphaned and unexplainable code within organizations. I'm currently, right now, dealing with that problem with a customer - only they got their unmaintainable code thanks to south asian contracting firms. Most of the jobs that genAI is going to replace first is that tier of contracting job, since the work product is equivalent and there's much less needful that needs doing.

Agreed. Agentic AI is going to be a huge deal across all verticals IMO.

I heard on a podcast an amazing interview with one of these AI founders and thought it was actually quite empathetic (and honest) in that he essentially said, "the job displacement narrative is totally reasonable and you are a fool or worse yet a liar if you are peddling it as fantasy for everyone. If you are a call center in Manilla or a paralegal, there is a real conversation that needs to be had about job displacement and it's not naive, but it dismisses the second order of facts that there are huge upsides and positives as well".

Add into that, not every red nickel saved is going straight to shareholders. Those firms will be out-innovated, out-competed and out-manuevered. A lot of that savings is going into strategic investements.

Last thing and unrelated, but I thought it was interesting the detail around the court ruling from a couple days ago:

Quote

The hottest business in 2025? A used bookstore near an AI office. Anthropic, maker of Claude and subject of endless copyright lawsuits like every other AI leader, had one model-training tactic revealed in court this week: they spent “many millions of dollars” buying up used books, unbinding them, cutting pages, scanning them, and ultimately binning them. The judge ruled it was fair use to digitize the library full of print books the company had purchased. What the judge wasn’t on board with: Anthropic also pirated 7m+ books. 

 

Posted (edited)
47 minutes ago, Vegas64 said:

I'm a retired history teacher who has too much time on his hands and super interested in technology from an angel investor perspective and from  a sociological perspective (just bought Daniel Susskind's new book, A World Without Work).

As someone who is not technical, I've found AI to be very liberating. It's not quite (and never will be) something that levels the playing field with someone like you who is very technical, but the number of semantic agents, vibe coding, and AI-enabling tools is very democratizing. 

There are very few real transformational periods to live through. PC revolution and mobile were obviously in our lifetime. AI is happening now and it's messy and painful but also beautiful and glorious.

Yeah, your perspective might be different from where I sit, and vice versa.

I think one issue at play that makes an analysis from a historical perspective insufficient is that the pace and scale of disruptive change is accelerating at an exponential pace, while the general cognitive ability of the human brain is not.  At the very least, technological advancements can far outpace evolutionary advancement.  How humans adapted to the scope and velocity of technological disruption in say the 15th century when scribes started being replaced by the printing press (a process that took over a century) is not necessarily applicable to how humans will adapt to the far greater velocity/magnitude of change that AI disruption will bring in the 21st century.

But do I think AI is the end of the world?  Absolutely not.  Life will certainly go on.

Do I think it could be the nail in the coffin of the period of democratized wealth enjoyed by most in the U.S. in the 2nd half of the 20th and early 21st century?  A period you enjoyed to be in the position you are today as a retired angel investor intrigued by AI? Absolutely.

I am not naive enough to think the AI genie is going back in the bottle.  What I hope happens is that we collectively choose to ensure the investor/board/executive class has to take into account the rest of society in their plans revolutionize the world with AI.  So I hope the attempted ban on AI regulation fails.  I don't think its in our best interests for the government to abdicate any ability to ensure something other than profit motive is governing the powers that be.

Edited by Goredho
  • Like 1
Posted
4 minutes ago, Goredho said:
  • I think one issue at play that makes an analysis from a historical perspective insufficient is that the pace and scale of disruptive change is accelerating at an exponential pace, while the general cognitive ability of the human brain is not.  At the very least, technological advancements can far outpace evolutionary advancement.  How humans adapted to the scope and velocity of technological disruption in say the 15th century when scribes started being replaced by the printing press (a process that took over a century) is not necessarily applicable to how humans will adapt to the far greater velocity/magnitude of change that AI disruption will bring in the 21st century.
  • But do I think AI is the end of the world?  Absolutely not.  Life will certainly go on.
  • Do I think it could be the nail in the coffin of the period of democratized wealth enjoyed by most in the U.S. in the 2nd half of the 20th and early 21st century?  A period you enjoyed to be in the position you are today as a retired angel investor intrigued by AI? Absolutely.

A couple of rejoinders, if I may:

I completely agree with your first bullet point (yes, I took the liberty). The velocity of the transformation is what is unprecedented and painful (and beautiful). The analogy I've seen is "it’s like we’re all accountants and Microsoft Excel came to be full maturity overnight". I think in such a massive disruption, there will be room for "new players to enter the chat" and new companies that become leaders and displace older ones who don't adapt or transform.

To your second point, of course. But the AI doomers that paint a dystopian view of life I think are off base. Not that they can't be correct but that they are so sure that they are correct.

Third point: It's been argued, I think that the democratized wealth in a capitalistic democratic republic like ours was manufactured and artificially created coming out of WWII, and was the exception not the norm. I think we are entering a period that could be argued as even more democratic than before in that we are deomcratizing access to expertise across the board.

Information disparity via exclusive circles, schooling, families with knowledge and lessons to pass on to their kids (e.g. the things we would identify as privilege and which largely acts as a prerequisite for things like income despairity) will be leveled. Most people around the world will have access for free or for cheap to good legal advice, good medical advice, personalized tutors, business advice, whatever. Knowledge that used to be extremely expensive and exclusive. Huge.

 

Posted
12 minutes ago, Goredho said:

I think one issue at play that makes an analysis from a historical perspective insufficient is that the pace and scale of disruptive change is accelerating at an exponential pace, while the general cognitive ability of the human brain is not.  At the very least, technological advancements can far outpace evolutionary advancement.  How humans adapted to the scope and velocity of technological disruption in say the 15th century when scribes started being replaced by the printing press (a process that took over a century) is not necessarily applicable to how humans will adapt to the far greater velocity/magnitude of change that AI disruption will bring in the 21st century.

Agree with this, but I go back to the earlier and much more recent example of the histrionics about the advent of the internet.  It spawned many more multiples of jobs in fields never before imagined than it eliminated.  This is only a few decades past.  

  • Hook 'Em 1
Posted
1 hour ago, Brisketexan said:

We are going to do more to make sure that no teenager can see tits on the internet

AI is pretty good at some things (loosely walking through technical problems, writing grammatically correct yet asinine prose, cooking advice) and terrible at other things (namely accuracy and truthfulness).

My uniformed guess is that this is a result of its training, largely internet writing. ChatGPT’s default tone is the same as Reddit posters.

With that in mind, and considering the history of the internet, AI will no doubt make access to more and weirder porn even more ubiquitous.

Considering AI’s disinterest in factual accuracy and its ability to confidently argue against reality (again, consider its training), I see us going one of two ways: 1.) learning to value factual learning and actually knowing stuff in your head again or 2.) losing our grip on reality. It’ll probably be some of both.

  • Like 1
  • Haha 1
Posted
12 minutes ago, Mole said:

AI is pretty good at some things (loosely walking through technical problems, writing grammatically correct yet asinine prose, cooking advice) and terrible at other things (namely accuracy and truthfulness).

My uniformed guess is that this is a result of its training, largely internet writing. ChatGPT’s default tone is the same as Reddit posters.

With that in mind, and considering the history of the internet, AI will no doubt make access to more and weirder porn even more ubiquitous.

Considering AI’s disinterest in factual accuracy and its ability to confidently argue against reality (again, consider its training), I see us going one of two ways: 1.) learning to value factual learning and actually knowing stuff in your head again or 2.) losing our grip on reality. It’ll probably be some of both.

As I understand it, the output is just as good as the data. A story as old as time with data and analytics (e.g. garbage in, garbage out). I'm not technical enough to understand the roadmap, but from what I can see there is already exponential advancements everytime a new model is released to eventually get to a point where it's pretty rock solid. Things like reasoning, model switching, RAG, MCP, and the like for today-- who knows what tomorrow will bring.

No alternative text description for this image

Posted
1 hour ago, Captainant said:

It's pretty remarkable what you can accomplish now with an agent to help with implementation and limited troubleshooting. With that said, it's also going to lead to a massive explosion in orphaned and unexplainable code within organizations. I'm currently, right now, dealing with that problem with a customer - only they got their unmaintainable code thanks to south asian contracting firms.

 

I’m in a similar boat where I’m at. To the point that on top of unfucking all that mess, it’s a full rewrite on the side as well.

The deeper we get into the contracted app, the worse it gets. To the point I can guarantee at least 80% of the code wasn’t tested.

Posted
20 minutes ago, Vegas64 said:

As I understand it, the output is just as good as the data. A story as old as time with data and analytics (e.g. garbage in, garbage out). I'm not technical enough to understand the roadmap, but from what I can see there is already exponential advancements everytime a new model is released to eventually get to a point where it's pretty rock solid. Things like reasoning, model switching, RAG, MCP, and the like for today-- who knows what tomorrow will bring.

No alternative text description for this image

Better data doesn't fix the inherent problems LLMs have in regards to hallucinations and reasoning. 

  • Hook 'Em 2
Posted
4 minutes ago, Dahobbs said:

Better data doesn't fix the inherent problems LLMs have in regards to hallucinations and reasoning. 

It helps with the factual part of it. I mentioned things like RAG for hallucinations, as well as SLMs, MCPs which, as limited as they are today, are just the beginning. I think it's foolish to think that we are not going to figure out how to get our colletive arms around providing a great user experience and reducing the hallucinations and bad factual answers.

Posted
41 minutes ago, Vegas64 said:

A couple of rejoinders, if I may:

I completely agree with your first bullet point (yes, I took the liberty). The velocity of the transformation is what is unprecedented and painful (and beautiful). The analogy I've seen is "it’s like we’re all accountants and Microsoft Excel came to be full maturity overnight". I think in such a massive disruption, there will be room for "new players to enter the chat" and new companies that become leaders and displace older ones who don't adapt or transform.

To your second point, of course. But the AI doomers that paint a dystopian view of life I think are off base. Not that they can't be correct but that they are so sure that they are correct.

Third point: It's been argued, I think that the democratized wealth in a capitalistic democratic republic like ours was manufactured and artificially created coming out of WWII, and was the exception not the norm. I think we are entering a period that could be argued as even more democratic than before in that we are deomcratizing access to expertise across the board.

Information disparity via exclusive circles, schooling, families with knowledge and lessons to pass on to their kids (e.g. the things we would identify as privilege and which largely acts as a prerequisite for things like income despairity) will be leveled. Most people around the world will have access for free or for cheap to good legal advice, good medical advice, personalized tutors, business advice, whatever. Knowledge that used to be extremely expensive and exclusive. Huge.

 

Yep, at this point it's all conjecture and theory crafting about what the actual effects will be.  The only thing we have to go by are the hints of intent from people like CEO Tamay Besiroglu who has created a business with a mission statement to "fully automate away all human labor" while citing the collective cost of world wages as his company's total market cap potential.  So, while it's cool you are excited about the possibilities, I'm wary.  So are a lot of others.  And given that no one knows definitively how things are going to shake out, I think it's asinine for the government to bar itself from any form of correction through regulation.

  • Hook 'Em 1
  • Like 1
Posted (edited)
7 minutes ago, Goredho said:

Yep, at this point it's all conjecture and theory crafting about what the actual effects will be.  The only thing we have to go by are the hints of intent from people like CEO Tamay Besiroglu who has created a business with a mission statement to "fully automate away all human labor" while citing the collective cost of world wages as his company's total market cap potential.  So, while it's cool you are excited about the possibilities, I'm wary.  So are a lot of others.  And given that no one knows definitively how things are going to shake out, I think it's asinine for the government to bar itself from any form of correction through regulation.

I get what you are saying in that this administration will not regulate federally so by default not allowing states to create their own regulation is allowing for no regulation. I think it's pragmatic to say we don't know and it's fair to be on either side of the issue is my only point.

Edited by Vegas64
  • Hook 'Em 1
Posted
8 minutes ago, Vegas64 said:

I get what you are saying in that this administration will not regulate federally so by default not allowing states to create their own regulation is allowing for no regulation. I think it's pragmatic to say we don't know and it's fair to be on either side of the issue is my only point.

Yeah, I certainly hope AI ushers in a golden age for the most people.  It's definitely not going away, and if anyone wants to take a principled stand against it, they better be financially independent.

Posted
33 minutes ago, Vegas64 said:

It helps with the factual part of it.

Not really, at least not without making your training set very specialized and limited (e.g., by removing all fictional or creative works). But then you couldn't get the scale needed for the more impressive capabilities seen with these models. You may be able to mitigate hallucinations to some degree, but the reality is that LLMs don't have any inherent way to identify "truth" or to reason. I think LLMs are going to ultimately end up finding their best use as a transformational layer between an input and another type of neural network or algorithm.   

  • Hook 'Em 3
  • Like 1
Posted (edited)
20 hours ago, Brisketexan said:

Apparently, we're all-in on a plan of "the beast will consume everything, and then begin consuming itself, complaining about the fact that it's hungry now that there's nothing left to eat."  Fucking brilliant.

I gotta admit, I'm truly amazed at how rapidly we've accelerated to across-the-board goals of "let's destroy society and humanity completely" all in the service of -- for a little while, at least -- funneling all remaining wealth to a few hundred oligarchs.  The short-sightedness is something to behold.

I just feel really, really, really fucking bad for my kids.

Re kids….
 

There’s nothing we can really do about the macro events but we can teach them to make AI a tool and not be collateral damage to the AI revolution

For me I think the answer is (1) economics and (2) quality of life.

economics - don’t be a worker, be an owner.  Own assets, income producing assets. I think this is self explanatory. 

quality of life - go analog where you can. Read paper books, do things in nature, talk to humans face to face. 

there is a third factor which is do not short cut knowledge and skill acquisition. AI helps once you are an expert, AI can help you become an expert but do not let AI be the expert.

to be honest, I’m one for three on this and I’ve got another one that may understand the economics part but no way the analog part.

but once you master these two then AI can remain a tool to help you accomplish both instead of it becoming your disease. 

I think those of us in our late 40s and 50s have a golden opportunity if we can use AI as a tool. The kids have a different challenge - acquiring native knowledge and skills while using AI as a tool instead of relying on AI as the expert itself. Both are opportunities. 

Edited by troph
  • Hook 'Em 1
  • Like 1
Posted (edited)

I used my own instance of chatGPT to help with an internal investigation in a workplace sexual harassment claim a few weeks back. I de-identified all parties, fed the witness statements and management reports into AI, asked for interview questions for the accused, recommendations that then formed the basis of an AI generated firm memo, a client memo, notice of termination, and email updates to the victim and witnesses.

now the most important part is I knew what the outcome should be but with that expertise AI saved me 4 or so hours and the client 4 or so hours and I did not do that crap work I hate (preparing interview questions, memos etc.). All drafts were reworked with AI 4-5 times and then human revisions at least twice. Factual inaccuracies had to be removed and reworked but in the end we had an 8-10 hour project reduced to 4-6 hours and I didn’t have to do any of the shit work. 

in that case it was pretty damn good. 

Edited by troph
Posted (edited)

I will add all of the lawyer focused and marketed solutions that I’ve seen so far suck though. Redline functions are absolute hot garbage right now - two reasons - redlining even basic documents requires understanding of intense factual scenarios, leverage and more important perceived leverage and lastly the ability to make sound judgment in light of those intense factual and subjective inquiries.  Most think redlining is like coding, it’s not, redlining documents, is more like arguing with your husband or wife than it is creating software code or rummaging through medical symptoms in aid of a differential diagnosis.

i don’t know how much time is needed for AI to develop that skill or if it ever will but from an output/results perspective it’s no where close. 

Edited by troph
  • Like 1
Posted
4 hours ago, HenryJames said:

If AI eliminates lawyers, I’m fully on board.

It’s one of the primary early stage objectives of authoritarianism so be careful what you wish for there. 

Posted
41 minutes ago, troph said:

It’s one of the primary early stage objectives of authoritarianism so be careful what you wish for there. 

As long as the attorneys don’t use their free time gained through AI to take over the world, I guess I’m cool with that.

Posted
52 minutes ago, Bevo said:

As long as the attorneys don’t use their free time gained through AI to take over the world, I guess I’m cool with that.

Pretty sure we'll just use it to post on Surly.

I have data on that hypothesis.

  • Haha 2

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.



×
×
  • Create New...