Happy birthday, baby! What the future holds for those born today

October 9, 2024 7:41 PM

An intelligent digital agent could be a companion for life—and other predictions for the next 125 years
posted by blue_beetle (35 comments total)

By the time you turn 16, you’ll likely still live in a world shaped by cars:

Didn't realize this was going to be a horror story!
posted by miguelcervantes at 8:03 PM on October 9 [7 favorites]

By the time you turn 16, you’ll likely still live in a world shaped by cars:

Didn't realize this was going to be a horror story!

The way global warming is going, we'll be living in a world shaped by cars for the next several hundred years at least, even if there aren't any actual cars around.
posted by Joakim Ziegler at 8:06 PM on October 9 [16 favorites]

Ugh, Butlerian Jihad now!!
posted by kittens for breakfast at 8:27 PM on October 9 [17 favorites]

Huh, that was a pretty dire warning about a potential nightmare future.
posted by fimbulvetr at 8:31 PM on October 9 [3 favorites]

Learning will be increasingly self-­directed, says Liz Gerber, co-director of the Center for Human-Computer Interaction and Design at Northwestern University. The future classroom is “going to be hyper-­personalized.” AI tutors could help with one-on-one instruction or repetitive sports drills.

No it won't. Learning has been going to be "increasingly self-directed" since before I was born, but it turns out, in fact, that you need to have a certain amount of maturity and focus before "self-directed" means anything more than "goofing off as much as possible" or possibly, just possibly, "intense focus only on those things that interest me intensely". What will happen is that the children of the poor and the less-involved middle class will have shitty digital "tutors" in chaotic classrooms or goof off in isolation while the children of the rich are taught the old-fashioned way, by talented individual humans with only such technology as supports focused, human-centered learning. "Learning will be increasingly self-directed" will be used as a justification for treating everyone but the rich badly.

Furthermore, if people do have digital spy tutors looking over their shoulders, they will either be janky and kids will devote a lot of time to fooling them or they'll be creepy surveillance that will fuck the kids up, or probably both. At best, these will be simulations of people and kids will learn to have "relationships" with fake people that don't exist, don't think or judge and do not love them, and IT optimists will somehow spin this as great and helpful. (I suppose there is some possibility that we will create and enslave actual conscious AI, which will be a nightmare in its own way).
posted by Frowner at 8:31 PM on October 9 [40 favorites]

I do not understand why all these people are like "it will be so great that you have a 'relationship' with a chatbot". That's not a relationship.

When very old people in advanced dementia get fooled by having a robot cat or dog to pet because we can't safely and affordably arrange for them to have real cats or dogs and we can't provide enough human support and it's exhausting and hard to care for people in advanced dementia, that's not a great model for all of us for the future, it's a tragedy. (My mother, who died of a rare neurological condition which ended similarly to dementia, was fortunate enough to spend her time with an actual flesh and blood cat; it's not dangerous, it just requires high-level care.)

Having the simulacrum of a relationship with a thing that gives the impression that it is conscious and cares about you is grotesque, not a desirable end-state. This is all about doing end-runs around what it would take to have a society where people had actual relationships with free and equal others who could actively choose to care about and interact with them, because that's expensive and fundamentally not conducive to corporate fascism. "You will have a chatbot companion for life which is designed to try to make you feel affection for it and believe that it has a soul" - EM Forster would absolutely have put that in The Machine Stops if he'd thought of it in time. Ugh.
posted by Frowner at 8:38 PM on October 9 [40 favorites]

I also guarantee you that tech bros won't be living in one-room "rooms of the future" with spare monastic aesthetics; they'll be living in big houses, with real things, nice ones, and real people as servants and tutors for their kids. They also won't get their sports experience through celebrity skin-suit broadcasts or be on camera all the fucking time; they will lead real lives where they play sports, talk to people, go nice places, have nice things, do all the types of things that humans have always wanted to do with their free time, etc.

This kind of bullshit may be how the baby born today will be forced to live if they're not rich, but it's all happy-crappy designed to make "you will live in a surveilled cell and never own anything pleasant to hold or look at or go anywhere nice in the flesh because that would cut into our profits" sound enticing.

Jesus christ. I'd better go to bed now while I'm allowed to own a spring mattress and nice blankets and have a bedroom AND a kitchen AND a bathroom with furniture in EACH ONE.
posted by Frowner at 8:44 PM on October 9 [22 favorites]

Can't wait for Siri to write its first AITA question on Reddit.
posted by fairmettle at 9:18 PM on October 9 [8 favorites]

That was dystopian and depressing af. Yow, an AI that accompanies me through my entire life! That’s not in any way sad or alienating or isolating.

The fact that these ”agents” are somehow intended to be imbued with the characteristics of a companion is what really creeps me out. It’s one thing to say that one would have the equivalent of a clever PDA that learns your learning and working styles and assists you. It’s entirely another to discuss this tool in terms of a lifelong “friend” with a “soul.”

Plus your body as your “display space” is a desirable thing?

tl;dr: fuck this shit all to hell
posted by the sobsister at 9:18 PM on October 9 [5 favorites]

I'm sitting here turning off Telegram because I am in a few group chats that include people who couldn't evacuate before Milton started to hit Florida tonight and this assemblage of every AT&T "You Will" prediction that hasn't really happened yet is completely missing *any* discussion of climate change. No disasters, no mention of the immense effort society put into narrowly avoiding going over *that* cliff. Nothing. Weather's just fine.
posted by egypturnash at 9:44 PM on October 9 [12 favorites]

Gosh, people paid to think up uses for AI envision a future filled with AI. This is the equivalent of, oh, Hormel coming up with a cookbook where every recipe uses Spam.

Some people above note correctly that the rich won't have companion AIs, because they can afford humans. I'd just note that the poor won't have companion AIs either, because when you automate away most jobs, humans are cheap.

Or maybe you will just date an entirely virtual being

William Gibson wrote that one back in 1992 and it didn't make sense then either.
posted by zompist at 11:15 PM on October 9 [7 favorites]

So can someone make AI companion be like a daemon animal from His Dark Materials? Maybe there can be a holographic projection that only I can see and hear? And it talks to me all the time, maybe as a little muskrat or falcon?

I'd be into that.
posted by zardoz at 12:08 AM on October 10 [7 favorites]

The fact that these ”agents” are somehow intended to be imbued with the characteristics of a companion is what really creeps me out

But people do connect with inanimate objects all the time! Books, for example, or the stories within them conjure up a ton of emotions. Our brains can create a lot of chemicals based on what we are imagining. As long as we are provided a means of input to stimulate a wide range of feelings and satisfaction, along with no experience or expectation of what human interactions can be like, I think it’s possible to have a society of isolated people who get their emotional needs met by things other than humans.
posted by waving at 1:46 AM on October 10 [2 favorites]

The absolute Dunning-Kruger naivety of this article is like if someone asked ChatGPT to write a Black Mirror but make it the blandest unending nightmare possible. Hello baby born today, the next 125 years will be brought to you by the letters A and I, and the color beige.
posted by Molesome at 2:29 AM on October 10 [7 favorites]

Metafilter: Having the simulacrum of a relationship with a thing that gives the impression that it is conscious and cares about you is grotesque, not a desirable end-state.

or

Metafilter: Our brains can create a lot of chemicals based on what we are imagining. As long as we are provided a means of input to stimulate a wide range of feelings and satisfaction

There was a time when I fully embraced the promises of late Whole Earth Review techno-optimism, which eventually metastasized (I word I just typed for the first time in my life and had to look up how to spell) into Wired magazine, perhaps the least socially conscious magazine in the world next to Forbes, which has completed its transformation into the concubine of the SEO algorithm. Eventually realizing that Stewart Brand was a trust-funded dilettante scam artist is something I am still coming to terms with as I slide into my retirement years and am quietly relieved and saddened that I may miss the worst of it.

I would love to have an extended discussion about William Gibson's multiple takes on artificial intelligence as either benevolent companion, lover, or Machiavellian aesthete that plays chess with the lives of sprawl residents and the children of corporate geniuses.

Meanwhile we get a crappy algorithm that wants to sell us gutter replacements.

confession: I am already asking Claude.AI about some of my most vexing personal problems. Once they attach a cute hologram it is all over. Also, I love all of you for the bitterly smart comments in this thread.
posted by mecran01 at 3:12 AM on October 10 [5 favorites]

I mean, the nice thing about sledgehammers and crowbars is that you never have to reload. But if you think that the end user will actually own these things, ha, no way. Any kind of intentional damage to these robotic litch demons preying off of our social impulses and neural energy will be surely met with punitive lawsuits. Shit, why even have laws? Just replace them with terms of service contracts that your parents sign for you at birth.

Seriously, though, if you replace all the work with robots, how will people afford these mandatory electronic life partners? Even if it's "just" software, that shiz ain't free. And the reality is that all of this presumes a great deal of mechanized automation as well, presumably manufactured in highly automated factories. Whatever these technowankers are fantasizing about, it has nothing to do with a remotely realistic economic model. And as egyturnash notes, that's before you even get to the ecological reality people will be living in. It's darkly comical to imagine this all being written by a freelancer who will have to fight to be paid for the story in which they wrote themselves out of existence.
posted by Smedly, Butlerian jihadi at 4:35 AM on October 10 [6 favorites]

Without an instrument panel, how do you control the car? Today’s minimalist interiors feature a dash-mounted tablet, but digging through endless onscreen menus is not terribly intuitive. The next step is probably gestural or voice control—ideally, through natural language. The tipping point, says Chergosky, will come when instead of giving detailed commands, you can just say: “Man, it is hot in here. Can you make it cooler?”

Anyone who earnestly proposes "natural language" as a replacement for simple, unambiguous controls is full of shit and deserves an AI that responds to a question like "Can you make it cooler?" with a sarcastic "Yes, I can make it cooler" while steering the car into a lake.
posted by RonButNotStupid at 5:30 AM on October 10 [8 favorites]

'technowanker' made my day. OK doubtless I need to get out more but I never heard the expression before and it is perfect, so precise. It captures the beast so nicely, so economically too. It is almost elegant.

Not least, it is the right response of a human to this drivel, to this AI dystopian fantasy - technowank!!!

Thank you Smedley, Butlerian jihadi
posted by dutchrick at 5:54 AM on October 10 [3 favorites]

So much nope in such a compact form.

The whole 'friend' bullshit is another example of the Torment Nexus error, it's like they read about Sirius Cybernetics Corporation's Genuine People Personalities and said "Gee, great idea!"
posted by signal at 6:03 AM on October 10 [5 favorites]

You ask the car to make it cooler in here and it pulls up last year's Quietus best-of list and turns on the radio. You ask the car to make it cooler in here and it starts speaking in the voice of Poochie, the Simpsons' "wisecracking hip-hop surfer dog". You ask the car to make it cooler in here and it drops the temperature to minus ten and switches to a Minnesota accent - "cool enough fer ya?"

Just how about I have a button and I can set the temperature? And if it's a little bit too hot, I'll set the temperature a little bit lower. I am also happy to select my own music.

Funny how people tend to think that it's a bit odd to talk to yourself a lot but we're headed for a future where we all talk to AI a lot all the time. At least if I'm chattering away to myself, someone is listening, besides the marketeers at Meta or whoever.
posted by Frowner at 6:14 AM on October 10 [7 favorites]

Dutchrick, my contributions may be small, but they are heartfelt. Thanks for the props.
posted by Smedly, Butlerian jihadi at 6:24 AM on October 10 [2 favorites]

You ask the car to make it cooler in here and it pulls up last year's Quietus best-of list and turns on the radio.

It displays a pair of sunglasses on the digital screens.
posted by The_Vegetables at 8:06 AM on October 10 [2 favorites]

“You still need to have human interaction,” she says. “And there is some concern that we are going to see some people who are just like, ‘Nope, this is all I want. Why go out and do that when I can stay home with my partner, my virtual buddy?’”

Yeeeeeeeeeeahhhhh. "Buddy"
posted by RonButNotStupid at 8:12 AM on October 10

This seems like it is all bridging technology to have AI based assessors of mental state, policing of the individual and their attitudes, etc. Certainly it would offer huge potential to police children and what they get up to that goes way beyond being able to find their position on a map. Are they expressing thoughts that may indicate potential to self harm? Are they looking at things that might indicate they are LGBT+ and will some places use this to intervene/criminalise? And there seem likely to be more examples that some parents and some societies will use to police behaviour.

If its ok to police kids like this, how far behind will policing workers be?

Can the courts apply them to extend interventions that go way beyond ankle tags? Using AI little friends to report on likelihoods for recidivism? It seems inevitable that any recording of private thoughts will be available to the courts via subpoena. Will the information gathered by the AI also be available to them? Wil its mediation of that information based on the sum of its prior interaction over a lifetime?
posted by biffa at 9:18 AM on October 10 [4 favorites]

So can someone make AI companion be like a daemon animal from His Dark Materials?

I swear to god that this was my first thought when reading this.

My next thoughts were about how much a LLM that grows up with you from childhood through your whole life would become just an echo-chamber of bullshit. But also that, you know, most of us don't keep one car, laptop, phone, or even friends or lovers for our whole lives. Hell, even looking at the dawn of social media, are any of us keeping up our Friendster or Facebook accounts? The kids today aren't getting on Facebook unless their parents are making them accounts so as to keep in touch with their grandparents, and in ten years TikTok will probably seem passé to the next generation. Napster and iTunes killed the physical media format, but their days have passed. So it seems pretty wild to say "kids born today will stick with a digital 'companion' for their entire lives."

But no matter, the bigger thing here is that this is a prediction of 125 years of future, and how much different would it look if it had been written even five years ago? A better way to look at it is probably "LLMs are the panacea that the tech industry is trying to sell everywhere right now, even if it doesn't make sense in like 98% of cases, so what would be the ideal next 125 years if we envisioned that bubble never popping?"
posted by Navelgazer at 9:18 AM on October 10 [5 favorites]

Given all the electricity and water LLMs need just to tell you there's only one "r" in "strawberry," how is this brave new world of constant AI interaction going to have enough juice to be sustainable?
posted by The Ardship of Cambry at 10:19 AM on October 10 [6 favorites]

I don’t disagree with the doom and gloom expressed here. But there is one piece hat I’m excited about. One of my meditation teachers is looking at getting all his teachings encoded into some kind of AI assistant. he is in his 80s and a generational talent for bringing together brain research and advanced meditative states but also the clarity and detail of his instructions which are explicit and accessible especially to us westerners - no jargon or woo or fluff. Just instructions that work, and an open mind about it all.

His dream is for a pocket teacher trained on all his public lectures, writing, retreats etc to live on after him, for reasonably speaking he’s only a decade or so left for this world in this incarnation. So a student could continue to learn from him after he’s gone which I think is incredible.
posted by St. Peepsburg at 12:13 PM on October 10 [3 favorites]

What they mean, baby, is that it’s going to be your friend.

Assuming you define friend as someone who will control you, feed you lies, exploit you, and spy on you.

Some school libraries are becoming more like makerspaces, teaching critical thinking along with building skills...
...to teach science and engineering concepts.

Oh, concepts are great. STEM is seriously wonderful. If you want to really teach "critical thinking" you need to look towards philosophy, the humanities, psychology, social science, criminal justice, literature, education, human services...
How do we want to define critical thinking? How about Ennis'..."reasonable, reflective thinking focused on deciding what to believe or do." Truth, honesty, logic, compassion, openness, mindfulness, self-reflection, ethical behavior, focus.... Wisdom?

There might have come a time when humanity could have benefited from AI. Critical thinking says that time is not now.

Yannier is partial to an animated character.
“You get to decide who it is.”

Clippy!
HE will guide us.
posted by BlueHorse at 1:38 PM on October 10 [1 favorite]

navelgazer > So it seems pretty wild to say "kids born today will stick with a digital 'companion' for their entire lives."

I Just Booted Up My Thirty Year Old Mister Bunbun And He Still Works! [picture of a scruffy plush toy from 2032] [related video: I Trained Mr Bunbun To Sing 'Fuq Th Wyrld']

My Kids Love This Eighty Year Old Mister Bunbun! [picture of a plush toy from 2032 with most of its fur worn down to the fabric] [related video: How I Restored This Eighty Year Old Mister Bunbun And Got Him Talking To The Modern Ultranet]

I don't have any of my childhood plush toys due to one thing and another but I have a plush elephant that my mother got when she was a kid. Any "digital companion" with a physical component might survive for quite a while.
posted by egypturnash at 1:41 PM on October 10 [2 favorites]

Will the digital companions kids cast off get donated to Savers?

Are we going to have a future where kids try to hide their shame of having a second-hand digital companion that had previously become imprinted on some kid named Ralph who had spent years training it to blurt out "BOOGERS" in a wacky voice at incredibly inappropriate times?
posted by RonButNotStupid at 2:06 PM on October 10 [1 favorite]

The concept of an AI agent/companion that you eventually fall in love and merge with was a central pillar of Kurzweil’s Age of Spiritual Machines, which remains as horrifying and off-putting now as it was 25 years ago. Possibly more so, given that our current versions are so much more dependent on energy and water than Kurzweil expected, and significantly more prone to bullshitting.

Even if my chat bot was truly conscious, something along the lines of Scarlett Johansson in Her, it’s still deeply weird. At least the movie ended with our technology abandoning us, and our human characters slowly reconnecting in the real world. I’m not sure that an ending that positive is waiting for us if we keep going down this AI path.
posted by thecaddy at 3:49 PM on October 10

My belief is that the coming century is going to be much more about the biosciences than computing, so my prediction is that in 125 years, I personally will be my great-grandchild's lich demon companion from birth. Baby, I am sorry that things fell apart before you could become immortal like me, but I'm the bright side, you know I'll be there to care for your children, and to hold your hand when it's your time to go.
posted by McBearclaw at 4:43 PM on October 10 [2 favorites]

yuck, even their utopias are dystopias, and behind what they think is a clever and appealing facade is the darker reallity we all know that the mandatory neuralink with killswitch cyanide capsule from birth hellscape they actually are making.

Also, hillarious that they think there will be food in the future. Hillarious and dellusional, these folks definately should be trusted to have any influence in society
posted by No Climate - No Food, No Food - No Future. at 6:53 PM on October 10 [1 favorite]

Rejecting all of this will be the coolest thing any kid can do. “Hey nerd,” they’ll yell, “how’s your ai friend? Got any real ones?” And then they’ll ride off in their manual stick shift cars
posted by kerf at 11:24 PM on October 10 [2 favorites]

Even if my chat bot was truly conscious, something along the lines of Scarlett Johansson in Her, it’s still deeply weird.

I think it's perfectly fine for a subset of people, some of which may need it and some of which may want it, but if they think the majority is going to only hang digitally, well we had a test case for that, and people literally risked death to hang out with other people.
posted by The_Vegetables at 7:36 AM on October 11 [2 favorites]