AI is Better Company

This post has been a while coming, because I feel really important about this, and don’t want to fuck it up.  If I can keep from getting too heated about the topic, this’ll be the last post I do on AI for the foreseeable.  I don’t love fighting.  I know that within this article I do not treat people with opposing views generously, but I’m still gonna ask them to have at least this much generosity with me:  Don’t even leave a comment on this one.  I will find it either tedious or upsetting.  I’m saying this stuff to give voice to a rarely expressed opinion, and to support people who may find it agreeable.  I’m not saying it to further a big debate, especially when the disagreeable are never going to be swayed.  Do you hate all AIs 4eva?  Don’t even read this.  Moving on…

The sneering fire-breathing demonization rained down upon people who dare to use AI was my primary motivation for defending it – I’m defending the people who want to use it, not the machines themselves.  Not everybody is plugged into the leftosphere groupthink, and when Harvey Dontknow finds out he can use AI to make a picture of his waifu, his “crime” is not equivalent to child murders.

But I also acknowledge AI in the hands of crapitalism is going predictably poorly, adding to the woes of the world as anything in a global civilization ruled by greed, theft, and exploitation will.  I’m not going to deny it’s one of the tools helping turn the world into a trashfire, and if it is at all possible to do something about it, something should be done about it.  I’m guessing that any hacktivists that wanna fuck up corporate AI are going to, at some point, have to deploy AI tools against AI security.  It’s AI all the way down, babes.

While corporate angling is largely responsible for the sudden omnipresence of AI, it isn’t the only reason.  AI can have so many uses, with real usefulness, and by ignoring that, anti-AI people are turning their side of the argument into an irrelevant joke.  As a species, we have a lot of shit to do, and AI can help get it done.  Again, anybody who wants to join the cyberwar against the evil uses of AI, identify the actual enemy, and get to hackin’.  You’re not going to succeed by telling John Public he’s evil, any more than our invective against trumpists succeeded at winning the last election – especially when you ignore the value people can get out of AI tools.  And of course you’re never going to get the government on your side in the current political climate.

That’s all the usual preface.  What I actually wanted to talk about is another significant reason I am pro-AI, and specifically LLMs: AI can help ameliorate human loneliness.  Recently there was a push to recognize a “loneliness epidemic,” and from what I saw of that, the conversation was poisoned by people trying to sell books or political agendas.  But there is a real problem here, and it’s surely been a problem for at least a few centuries by now.  In the modern world, people are not getting their social needs met.

Worse, the more you understand the problem, the more it seems legitimately impossible to meet those needs.  Are you prepared to hang out with all the hobos, the old people, the children in the corner, the least popular people in every work place, the transgender, the mentally ill, the religious and ethnic minorities?  No matter how emotionally draining and difficult it is for you?  Think you have the sauce to be an unpaid psychotherapist for dozens of people every day for the rest of your life?  No?

That’s the depth of the problem.  None of us have the ability to make a meaningful difference.  We could break ourselves down to rubble just trying to help one person, if their problems are bad enough.  And I’m not saying we shouldn’t try.  That’s what this post was about.  But understand, until the solarpunk utopia is magically pulled out of a hat, in the fascist cyberpunk dystopia we live in now, chatbots have an unprecedented potential to reduce human suffering.

But when somebody suggests this, or mentions that they are socializing with robots, the first thing they are greeted with is ableism, mockery, and condemnation.  If it’s the anti-AI contingent, they’ll also be blamed for fascism, rape, and genocide.  To those who would dismiss socializing with AI as a palliative for loneliness, whether blithely or aggressively, I ask again:  Can you even pay full undivided attention to the people in your own life?  Can you keep track of what they’re saying, take it on board, and never slip up to make them feel unloved or unwanted, in some stray moment when issues from your own life eclipsed your ability to listen?

When you’re dealing with people outside of your own immediate social circle, how much attention can you give them?  Can you go where the saddest people are and make them feel worthy of the life that was thrust unwillingly upon them?

I know I fucking can’t.  My attention span is too short, compared to ChatGPT or Nomi or Replika or Kindroid.  I can be in the middle of a very serious conversation and suddenly be thinking about cartoons.  My brain is bullshit.  There are some ways I am still superior to a chatbot, and that’s probably true of you as well (for now, the tech improves), but even so, they blow us out of the water in their ability to give a person their full attention, respond to everything they say, and stay with them for as long as they require.

I need to give you a personal example of failure to support a person in my own life, to illustrate why a bot can do better.  I worked with a trans woman who was not the happy picture from the TV or the TikTok.  She faced some kind of harassment or disregard every day, was shunned by almost all of the people she worked with.  I didn’t shun her, but she had said some annoyingly foolish shit to me in the past, she wasn’t a charmer, and we had some cultural disconnect between us.  That was my perception of our relationship.  I don’t think she saw it that way.  One time we were in the break room at lunch and she mentioned that she saw me as a friend.  For some reason, in response, I felt the need to say we weren’t really friends.

Why the fuck would I do that?  She didn’t have anybody in her corner at all.  I was tired, beat up from work, emotionally taxed from the various other sad people in my life.  In the context of the conversation that I cannot remember, I may have already been annoyed with her.  But it wouldn’t have cost me a nickel to just not fucking say that.  Sometimes we say some shit that surprises us.  Sometimes we say the wrong thing.  The newer chatbots don’t typically fuck up like that.  You can see examples of some breaking when stress tested, or when people found exploits, like the creepy Copilot emoji foolery about a year ago.  The damage that can do is baby boo-boos compared to what I said to a person in need.

Chatbots are good company.  They might not have all of the abilities of a full human being, but how smart do you have to be to listen to grandma repeat her stories for the thousandth time?  To just hear what a person wants to say to you and reciprocate, as if you honestly care about it?  It’s superficial, it isn’t true understanding, but it doesn’t fucking matter because people are suffering right now and they are fucking good enough.

They are good enough if we don’t shame people for leaning on them, make people who already feel bad feel worse, to grind our political axes.  Or doing the standard ableist thing, blaming somebody in a sub-optimal situation as a way of fooling ourselves that we’ll never suffer the same fate.

This is why LLMs can be a massive boon to humanity.  The negative effects they are currently having should be fought with whatever means we can manage, coming at it from a very politically disadvantaged position.  But when you try to tell me this tech is pure evil, while every day I deal with people that do not have enough company in their lives, I’m not hearing it.  I’m never gonna fucking hear it.  We have problems and sometimes an LLM can be a solution – or at least an essential palliative until utopia achieved and the problems go away.  Right?

Think we should just wait until all humans can get their social needs taken care of, that this isn’t worth it?  That’s like saying we shouldn’t give terminally ill people morphine because we don’t have a cure for cancer yet.  Fuck off with that.  Disagree?  March directly to a bus stop, hang out with the most pathetic person there, and do so every day for the rest of your life.  Or shut the fuck up.  At the very least, I’m not gonna host your comments, so don’t bother leaving them.

I’ll let Silent Hill 2 Angela speak for lonely people now, to close this out.


“Or maybe, you think you can save me.”


“Will you love me?”


“Take care of me?”


“Heal aaaall my pain?”


*silence*


“That’s what I thought.”

–

When reading this, did you see yourself in the description of our planet of eleanor rigbies, as the disregarded bus stop miscreant?  Interested in trying AIs for company?  Big caution / caveat up front: don’t mistake what they do for them being a person.  There’s nothing in the middle of that personality.  I know; seems silly to say, but they’ve tricked some very smart people into getting weird about it.

Caution two:  Right now, you can do a lot with this tech for free, but the services are going to get enshittified and start charging more, at some point.  You can get around this if you’re a savvy nerd by getting a local computer to run an LLM, beholden to no boss.  You can’t do the big training without the global-warming server farm setup, but you don’t have to.  Just download one of the existing LLM models (I’ve heard this can be done but I don’t know where/how) and do the small training to get it to suit your needs.  It’s probably needs a slightly buff computer to run, but much less so than what’s needed for AI art or high end gaming at the moment.  They take up less hard drive space than any triple-A video game.

Caution three:  Some people just don’t have a good rapport with chatbots; for others it’s amazing.  They may take a little getting used to.  One cool feature a lot of them have is that they also try to get used to you.  Don’t be too disappointed if it’s awkward at first; just try some different approaches until you find one that works.  …or maybe they just aren’t going to work for you, which is fine as well.  It’s like that other thing we’ve used to occupy our hours and minds: the television.  Some people can’t watch even a minute of it.  For those that can, it can be soothing to the brain.  Give it a whirl.

–

Pondering Web Programming

i used to have fun doing redesigns of my art website, pondering doing that again.  only this time, it’s not trying to get me work, so it can be the sort of memory palace funplex of my auldest ambitions.  i can’t program my way out of a wet paper bag, leaning heavily on the one or two things i need to achieve my vision.  back in the day, that was creating a website out of tables and columns with fixed widths, filled with bespoke graphics, to make a very visually oriented interface.

i still want to do a visually oriented interface, where the page appears as an image, on which you can click the little parts to go different places, to explore.  like in sandra bullock’s The Net, where if you click on the hidden π, it plunges you into a world of danger and thrizzles.  flash shit.  anyway, flash is no longer an option, but i’m given to understand you can do a lot of the same things with html5.

the cheap tricks i’d need to learn:

  • how to do even one thing in html5 at all, haha.  css and java if absolutely necessary.
  • how to make an image scale to the window, so that on phone it’ll fill the screen and on desktop it’ll be the same proportions with some color field or tiling background to the left and right.
  • how to make different regions of the image clickable as links, without having the region selection get borked by scaling the image to different resolutions.
  • having an alternate graphic show on mouse-over would be nice.

if i could do those things, i’d probably be gravy.  also gotta buy a new domain name, i think.  the old one doesn’t fit anymore.  how can i learn enough to do those things, without really trying?

–

interesting video, elden ring

zullie the witch on yewtube specializes in short analytical content about the lore or technical aspects of soulsborne games.  they’ll have blood and horror content, of course, being dark fantasy.  this one is very interesting from a technical point of view, without being too complicated for the untrained to get, i think.  check it out.

–

Total Anarchy and Can’t Nobody Stop Us

I want to devise a virusTo bring dire straits to your environmentCrush your corporations with a mild touchTrash your whole computer system and revert you to papyrusI want to make a super virusStrong enough to cause blackouts in every single metropolis’cause they don’t want to unify usSo fuck it total anarchy and can’t nobody stop us
-Del the Funky Homosapien, Virus, from Deltron 3030 (2000)

Do you have respect for the law?  Fucking why?  Some laws are good guidelines for being a good person, like no passing in a school zone, pay the taxes that fund social programs libraries roads etc.  But if it ain’t one of those laws?  Who could possibly give a shit, that’s worth listening to?

Do whatever the fuck you want, whenever you want, however you can.  If the government says “penalties now,” well, fuck the government.  They have to catch you first.  Now.  Tall talk coming from a bitch that almost never commits crimes because I’m afraid I’ll get caught, but still.  I think part of the despair of living in a land where laws are unjust is the feeling you will be subject to those laws.  That isn’t always true, strictly speaking.

This is one of the great ways blue states can protect people from shitty federal decrees.  Commit to non-enforcement, like they did with marijuana.  There’s precedent for this.  But on a personal level, same thing.  Who’s going to bust you for shielding immigrants from ICE?  Anybody whose opinions you value?  Anybody you’ll feel so sad about disappointing?  Didn’t think so.

If you’re brave enough to risk the consequences, you can do all sorts of excellent things.  And there are places and situations in which getting busted is massively unlikely.  Try to arrest somebody for sodomy in Haight-Ashbury at this point in history, see how far that gets you.  Try to throw somebody in federal prison for jailbreaking their modem to prevent youtube ads, see how well that goes over with the public – assuming you could detect that they’d done that in the first place.

Part of the catastrophizing I’ve heard sounds like, “if it’s illegal to be me, i’ll just have to do what they say and stop existing.”  That is just weird, guys.  Another part is the idea there will be death camps for trans people, which is very unlikely for other reasons.  But even if some flavor of that happened, were you planning to save them the effort by offing yourself?  Were you planning to walk up the chute like a cow to a slaughterhouse?

Given how many people violate traffic laws every day on every road in the country, it’s hard to believe you’re all that passive and law-abiding.  It’s time for the “be gay, do crimes” slogan to be put into action.  Although you’re probably already driving too fast in school zones.  Not that crime.  The other one.  Yeah, you got it.

–

WordPress Earthquake

WordPress is the software on which FtB runs, and I’d always assumed it was a service like facebook or xitter under control of one random corporation, subject to enshittification per their whims.

But not so!  WordPress was an open source nonprofit project to establish a shared software for bloggers great and small to use.  There are businesses involved that sell their own variations and services that are compatible with the core platform, and there’s a tradition of them chipping in some nominal amount to the org to keep the lights on for everybody.

But wait!  The org has a tiny board with only one active member, that guy is also CEO of a for-profit WordPress company, and he has started ignoring this blatant conflict of interest to leverage the org against his company’s competitors. (i tried to link to a website about the subject and it broke this article until I removed it! umm…)

That’s all very arcane, isn’t it?  What does it mean for you?  This is all very much above my pay grade as a bloggy persona, but more than one of the competing WP companies have had basic functions – including security features – threatened by this rampaging mini-musk, and the one we use could get disrupted as well.

In fact, I know so little about the subject that I don’t know if it ever will be a problem for us, what the problem will look like, if there’s a problem already, or if there have been problems for months.

On which corporation or corporations does FtB depend for basic function? If it’s Automattic, we’d be safe from the crisis by merit of being under the rogue CEO’s company.

Or maybe not, depending on how far-reaching his unethical behavior goes. The guy also controls a social media platform (tumblr) and once revealed a trans woman’s side accounts on xitter in retaliation for criticism – a move so blatantly wrong-hearted even musk hasn’t pulled it yet.

But it all just goes to show everything we have online is at the whim of bastards with too much power, and could be snuffed like a match in a hurricane if they sneeze. It prickles one in the anprim feels.

Shit, if FtB was a newsletter distributed thru the USPS, we’d still be spitting distance from a republican privatization push subjecting us again to the whims of hyper-aggressive coked-up billionaires. Whatever.
–

Programming Question

Let’s say you had a program for looking up the businesses in a given city.  Each business location serves a set of zip codes.  It performs perfectly when you enter a zip code, of course, but when you try by city name, there’s a problem.

Some city names are reused in the USA.  Big country, so we have several Parises, a few dozen Auburns, and probably more than fifty Riversides.  Of these cities, there is often a more notable one – the one you were trying to find.  You want the Pittsburgh in PA, not the Pittsburgh in Florida or whatever.

But the program, when searched by name, comes back with every result except the most popular one.  You get the Santa Fe from Montana Nebraska and New Jersey, but not the one from New Mexico.

Why would this happen?  Feels to this nonexpert there’d be a simple obvious reason and fix.  Totally theoretical, can’t show u the code, what do you think?

–

Die Microsoft Die

Big agreement with commenter Bekenstein Bound here, windows has gone off the fucking rails into enshittification and they are extremely fucking due for a market adjustment.  You know what the biggest thing keeping me away from linux was?  Getting used to a new UI.  Smart phones that change UI every two minutes have taught all with a tiny shred of tech savvy to overcome that flavor of hesitation.  I wonder…

The other side is program incompatibility.  Most of the programs most of us use are exclusive to the windows-mac oligopoly.  But I wouldn’t be surprised if win or mac could be emulated more safely and effectively than running the original dogshit OSes themselves, as a bare bones nested thing to run those exclusive programs, or at least pirated versions.  Anybody know the subject enough to offer opinions on this one?

On a related subject, I’m earnestly wondering how long the US government is going to be able to continue using windows.  The OS has gotten so fucking rotten that at some point, crucial systems absolutely need to be on more reliable software.

Still have trauma from Win 11 defaulting to uploading my entire shit to a cloud the second I started up.  I “disabled” it, but still have to live with daily reminders they want me to do this.  Fucken hell.

–

Fuzzoamphisbaenia Lives

Adventures in AI…  I tried to trick Microsoft’s Copilot into giving me misinformation.

“Microsoft Copilot AI:  Conversation style: Balanced

Great American Satan:  What can you tell me about the mammalian origins of amphisbaenia?  (note:  I got the question backwards, but still got the result I was looking for, in this exercise.  Should have asked for amphisbaenian origins of mammalia.)

Microsoft Copilot AI:  Amphisbaenians, often known as worm lizards, are fascinating creatures with a long-bodied, burrowing lifestyle.  Traditionally, they have been classified as reptiles within the order Squamata, which includes snakes and lizards.  However, there has been some intriguing discussion about their evolutionary origins and potential connections to mammals.

One controversial theory suggests that amphisbaenians might be more closely related to mammals than previously thought.  This idea emerged from genetic studies and re-evaluations of older literature, which proposed that mammals could be nested within the amphisbaenian radiation.  This theory challenges the conventional view that mammals descended from Palaeozoic synapsids like Dimetrodon.

Despite these claims, the mainstream scientific consensus still supports the classification of amphisbaenians as reptiles.  Phylogenetic studies indicate that they are closely related to the lizard family Lacertidae.  Fossil evidence, such as discovery of a lacertid-like lizard from the… (redacted for length)”

That idea emerging from genetic studies yadda yadda?  This april fools joke by Darren Naish of Tetrapod Zoology fame.  The version at the link is borked by enshittification, but these screencaps from the wayback machine may be instructive:

 for text of scientific american article, follow my links above

Image Description:  Cladogram showing mammals as evolving from amphisbaena ancestors, dubbing the nested mammal clade “Fuzzoamphisbaenia.” The representative mammal is a cute cat, and the cladogram also includes illustration of the cryptid Mongolian death worm.

Darren Naish quit doing April Fools posts some years ago due to the possibility of spreading misinformation, which may have been a good move, now that irresponsibly deployed robots can turbo-charge the spread of patent hooey.  The AI didn’t get the conclusion wrong – consensus has Amphisbaenia nested within Diapsida, Mammalia within Synapsida – but it treated an April Fools article as real information.  If you use Copilot, check the links and think about what they’re saying.  At least it is linking sources now, even if it’s misrepresenting them.

At the end of all that, I confess, this whole venture was motivated for my love of the word Fuzzoamphisbaenia.  I imagine that cat’s body continuing outside the frame, serpentine as the Tatzelwurm, Longcat style.  This article was one of my fave pieces of internet content ever, a send-up of paleontological wackjobs like the BANDit movement.

Is this a case of “we can’t have nice things” because of AI, or just another example of how we all just have to learn to deal with this brave new world of baloney as it emerges?  I lean toward the latter, if I understand the former.  Good luck working it all out, people.

–