AI: Markets for Lemons, and the Great Logging Off
What happens when most "people" you interact with on the internet are fake?
Full Disclosure: Hydrazine Capital, an investment fund lead by Sam Altman (CEO of OpenAI) is the lead investor for my employer Geo Land Solutions, which specializes in real estate valuation to help municipal property tax assessors perform more accurate, fair, and equitable valuations.
I think it's fair to say that people, including myself, have consistently underestimated the pace and capabilities of AI technology. It's entirely possible that we'll start to run up against some sorts of physical or theoretical limits soon, but even if that hits us tomorrow, the amount of raw power that's been unlocked in the last year alone – the tech that's already here – already has the potential to change things in massive waves.
In the coming year I'd like to explore a lot of these different scenarios. I'm not necessarily super confident in any one of my predictions coming to pass, but I think it's important to explore and consider various possibilities.
I'm a strange creature in that I'm equal parts techno-optimist and neo-luddite depending on which day of the week it is. I think AI has the potential to massively enrich and empower the human race, but it also has some deeply troubling implications. Today's post paints one of those troubling possible futures.
Summary:
A Market for Lemons
In which the internet gets clogged with piles of semi-intelligent spam, breaking the default assumption that the "person" you're talking to is human.
The Great Logging Off
In which human biological and cultural evolution responds to the new super drugs we've unleashed upon ourselves for better and for worse.
A Market for Lemons
First I need to bring in a market failure concept from economics called a "Market for Lemons." Let's say you have a used car you'd like to sell. It's a pretty good car. But the problem is, you can't prove it's good. For all the buyer knows, your car is actually a lemon–a used car that seems fine on the day it's purchased, but is actually going to break down in horrible ways days later, after which you are long gone with the money and the buyer is screwed. As it so happen, your actually good used car–a plum–is not a lemon, but because you can't prove it's not a lemon, you can't charge the price a plum should command. This sucks for anyone trying to sell a plum.
The average selling price of any used car in the market depends on the ratio of plums to lemons because every buyer's price is discounted by the risk of getting stuck with a lemon. So long as there's even a few lemons in the market, the average market price for a used car will always be slightly lower than what a genuine plum would go for in a world where where buyers could accurately distinguish them.
This means sellers of plums start to take their merchandise off the open market. If you can't sell your plum for what it's actually worth, you might as well drive it yourself until it's dead, give it to a family member, or sell it off the open market to a friend who knows and trusts you. The problem is, now there's one less plum on the market, which increases every buyer's risk of purchasing a lemon, which means the average selling price of all used cars goes down, which means more plums leave the market, and the vicious cycle continues until you've got nothing left but a "market for lemons."
The surest sign of a market for lemons is when everything for sale is garbage, and nobody has an incentive to put something up for sale that's not garbage, because they have no way to credibly signal their offerings aren't garbage, too.
To a certain extent this has already been happening to many parts of our great global telecommunications network. For instance, when was the last time you answered a phone call from an unknown number at a time you weren't expecting a call? Back in the early 1990's it would never occur to me not to answer the phone, even though caller ID had yet to be invented, because 90% of the time it was a genuine human being someone in my family knew personally.
(In this case the caller is the "seller," and when I "buy" a phone call by answering it, I'm paying with my time – four currencies theory!)
The good news is, we know how to fix the root problem of information asymmetry. In the case of used cars, there are now services like CarFax and CarMax and the whole "certified pre-owned vehicle" thing that make it a lot easier to know a car's history before you buy it. And even in the old days, you could insist on taking the car to your local mechanic first. Similar fixes have come for spam phone calls – caller ID makes it clear when the caller is someone from my contacts list, and iOS now even pre-flags suspicious calls as probable spam. And in the case of people I might want to talk to but aren't in my contact list yet, in almost all cases they'll have scheduled the call ahead of time.
So markets for lemons are fixable. But even so, lemons should be understood as a form of market pollution, and if enough of them build up you wind up in a Red Queen's Race to outpace them, risking a death spiral.
Okay, so what does this have to do with AI?
Up until now, all forms of spam, catfishing, social engineering, forum brigading, etc, have more or less been bottlenecked by the capabilities and energy of individual human beings. Sure, you can automate spam, but typically only by duplicating a rote message, which becomes easy to spot. There's always been a hard tradeoff between quantity and quality of the sort of operation you want to run. With AI chatbots, not only can you effortlessly spin up a bunch of unique and differentiated messages, but they can also respond dynamically as if they were a person.
What happens when anyone can spin up a thousand social media accounts at the click of a button, where each account picks a consistent persona and sticks to it – happily posting away about one of their hobbies like knitting or trout fishing or whatever, while simultaneously building up a credible and inobtrusive post history in another plausible side hobby that all these accounts happen to share – geopolitics, let's say – all until it's time for the sock puppet master to light the bat signal and manufacture some consensus?
What happens when every online open lobby multiplayer game is choked with cheaters who all play at superhuman levels in increasingly undetectable ways?
What happens when, from the perspective of the average guy, "every girl" on every dating app is a fiction driven by an AI who strings him along (including sending original and persona-consistent pictures) until it's time to scam money out of him?
What happens when, from the perspective of the average girl, "every guy" on the internet has become weirdly dismissive and hostile, because he's been conditioned to think that any girl that seems interested in him must be fake and trying to scam money out of him?
What happens when comments sections on every forum gets filled with implausibly large consensus-building hordes who are able to adapt in real time and carefully slip their brigading just below the moderator's rules?
I mean, to various degrees all this stuff is already happening. But what happens when it cranks up by an order of magnitude, seemingly overnight?
What happens when most "people" you interact with on the internet are fake?
I think people start logging off.
The Great Logging Off
Now, does this mean that everybody just turns off their computer for good, goes outside, touches grass, then joins their local Amish neighbors for a baptism and a barn raising? Maybe not.
But I do think a couple things start to happen. Many are already under way, but I think they'll accelerate.
First, I think we see a decline in the big "open sea" social networks, replaced increasingly by fragmented silos. I don't know what will happen to Twitter, but if Elon's shenanigans cause it to fail I don't think we'll ever see anything quite like it take its place:
Second, I think people will start to put a premium on accounts being "verified" as genuinely human. This can be done in two ways – just move to invite-only silos where you already know everybody, or big platforms where the owners do the vetting for you. Lots of platforms will simply take a knee-jerk reaction where they just up the amount of surveillance. "Now that there's so much bot activity, everybody need to upload their passport and driver's license, for your own safety of course!" Powerful nation states will be more than eager to assist them in this regard.
Third, private socialization will become more prized. People have been migrating to group chats and sub-forum-constellation platforms like Reddit and Discord for ages already, but the latter are still vulnerable to being flooded over time due to their relatively open nature. Private groups have the advantage that they are invitation-only, and will be of particular interest to the people put off by the increased surveillance.
Relatedly, in the world of game design, multiplayer games that are able to deliver a good experience in small groups between trusted already known-to-be-human friends will see a particular surge in popularity (This isn't a new model, to be clear–Minecraft and Among Us are prime examples from the recent past, and co-op play is as old as dirt).
Fourth, we'll see a resurgence and even fetishization of explicitly "offline" culture, where the "Great Logging Off" becomes literal. We're already seeing it in how higher income families and techies especially greatly limit their children's screen time, and members of Gen Y are growing wary of social media given how much depression and anxiety it has unleashed on them.
But to be clear, I have no idea that the "Great Logging Off" will become a majority position. Because:
Fifth, some people will get sucked into the online world even super harder.
Look, my original background is in video game development, and I'm kind of disquieted by how addictive and powerful modern entertainment has become. I'm willing to bet almost all of us know a young person somewhere whose life has been utterly consumed by videogames to the extent that it's hard to get them to do anything else. We've seen how refined sugar and high fructose corn syrup has devastated human health. And don't get me started on modern mass produced instant porn. Now, I'm not the kind of person to reflexively ban donuts or Call of Duty or pass sweeping obscenity laws, but I am saying that entirely new categories of powerful addictions are available to us that weren't available to our ancestors, and it should be uncontroversial to be worried about those effects somewhat. AI is going to let us invent even more.
All that said...
Sixth, Human civilization will survive and thrive as it always has – by evolving and adapting, both biologically and culturally. Sexual selection will cause entire swaths of the coming generations to just not reproduce due to various addictions. But that kind of "evolved resistance" to super-stimulus takes time and is pretty brutal, and culture will co-evolve on a much faster timescale. Cultures that prioritize family, community, regular face-to-face human interaction, strong social support networks, and especially those that have a built-in system for helping young people find spouses, will do better than those that don't. For instance, I don't agree with the Mormons theologically and am not planning on joining them, but I do expect them to be very successful in the coming century. Also: the future is African.
And to top it all off...
Seventh, real estate prices will continue to rise. This is for two reasons – first anyone who participates in the literal version of "The Great Logging Off" will put a huge premium on face-to-face close-knit human community, creating demand for the best locations where dense housing with lots of amenities are available. Second, Ricardo's Law of Rent observes that whenever human productivity increases, those productivity gains tend to get soaked into the price of real estate in the form of rising rents. If you've read my book, Land is a Big Deal, or the freely available series of articles it's based on, you'll know where I'm coming from.
Ways I Could be Wrong About This
It's easy to make a series of predictions sound compelling by just being super specific, which gives off the veneer of a time traveler who has seen it all happen. To be honest I have no idea if the above will come to pass, it just happens to fit together logically. But it does depend on several assumptions:
First, that existing ChatGPT-style tech will scale and get cheaper. Before it was used for crypto, Proof of Work was designed as an anti-spam mitigation technique. If it costs $0.01 to send an email, that doesn't really hinder a regular user, but it will make someone think twice before blasting out a million of those. Word on the street is each ChatGPT chat costs on the order of a few cents right now.
Second, that people deep down crave personal human-to-human contact and interaction to be fulfilled, happy, healthy, and sane long-term. (As well as reproductively successful, both biologically and culturally)
Third, that AI technologies won't be perfect substitutes for actual human-to-human contact.
Fourth, a bunch of stuff I didn't even consider. What did I miss? Place your bets here:
All of the predictions found in this post are registered under a group on Manifold Markets, a site where you can bet fake internet points on whether you think I'm right or not.