Amy Webb Launches 2026 Emerging Tech Trend Report

[applause] [applause] Good morning everyone. I'm Amy Webb. I am chief uh executive officer of Future Today Strategy Group. I'm also a professor of strategic foresight at the New York University Stern School of Business. And I would like to thank you all for being here today. We are gathered here today to celebrate and to remember the life of Tren report. [music] We were young with the world on rewind. Writing names in the margins [singing] of time. [music] Late nights, headlights, nowhere to go. Every answer was we just got to know. [music] We made plans we were

sure would come true. [singing] Drew a map. Every road led [music] to you. I can still hear the sound of your laugh. It [music and singing] was a promise that We thought would last. Every moment felt bigger back [music and singing] then with the future coming round the time whispering. Don't be afraid. Life will change but memories won't fade. Here's to the night you thought wouldn't [music] end all those what ifs and all we said. If this is goodbye, I'm all right with the [music] truth. We're better for what we've been through. I wouldn't trade

[music] a second. We had every page, every good, every bad. Now I'm closing the door. But I'm leaving [singing and music] the light for the moments that got us here tonight. [music] 18 years give or take a dream. Funny how fast [music] forever [singing] can leave. We chase meaning in headlines and signs. Tried to read every star in the sky. We held on through the [music] best and the worst. Every [singing] rewrote the first and I see now what I couldn't Before. [music] Change just means something more. Don't [music] wonder who's wrong or right. Just

able to felt so [music] alive. Every chapter has a last line. And this one's mine. Here's to the night we thought wouldn't end. All of those what [music] if we said if this is goodbye. I'm all right with the truth. We're better for what we've been through. [music] I wouldn't trade a second. We had every page, every Good, every bad. [music] Now I'm turning the chapter. Yeah. I'm ready to say it was beautiful. The days are on their [music] way. [music] [music] >> [applause] >> So, so friends, today we we are here to say goodbye

to Tren report who would have been 19 years old. And I know how strange that sounds, but grief doesn't Follow the rules of what's supposed to happen. Trend report was generous in the most literal way. It was published openly and freely. It asked for nothing and yet it gave us a common language across countries and industries, across business and filmmaking and art. It made conversations smarter, decisions better informed, and people more creative. Trend Report had a rare kind of authority. During a time that was just saturated with hot takes and loud Certainty, Trend Report didn't

need to shout. Trend report could take a flood of information, mountains of data, and turn all of that into a clear story you could act on. Its credibility wasn't branding. It was rigor. Hundreds of slides, a thousand slides at the end, grounded in real data. And that's why Trend report reached millions of readers every year. And we weren't alone. After we started publishing our trend report, many many more followed. Now across the Industry, across every industry, there are hundreds of trend reports published every year. Every consultancy, every agency, every brand, every think tank. We were

all just trying to make sense of the future, one trend at a time. So this is actually not just a eulogy for our trend report. It's for all trend reports. Look friends, the world is changing fast. A static PDF of trends provides information but it becomes stale Immediately because things are changing so fast in tech, in the economy, in policy and in my view in a lot of organizations these trend reports have become a crutch rather than a catalyst for strategy and for capital allocation. So, it made no sense to ignore all of that change

and to keep doing the exact same thing year after year because you can't reverse time. You cannot make the world the way that it used to be. So, we recognized our innovation for what it Had become, a once great idea worth destroying. Sometimes you got to burn down what you built and make way for what the future demands. And that is actually why we are here here today. So yes, technically this started off as a funeral. But saying goodbye on your own terms, that's also a cause for celebration. So we're going to make a transition

right now everybody. Because in saying goodbye, it is an opportunity to Celebrate to celebrate our deliberate act of creative destruction. So, I want you to get up. I want you to stand up right now. Get out of your seats. Get your cameras ready. And South by Southwest, I want you to open up those back doors. dude. [music] [music] [music] Heat. [music] >> [music] [music] >> Heat. [music] Heat. [music] Heat. [music] Heat. [music] [music] >> [music] [applause] [music] >> Heat. Heat. >> [music] >> Heat. Heat. [music] Heat. [music] [music] [music] [music] Heat. [cheering] Heat. Heat. [applause]

[music] [music] [music] >> [music] [music] [applause] [music] >> All right, now we can settle back in and get started. I want to friend thank my friend Dr. Cliff Krooms. uh for being here this morning with the Texas Longhorn Band. Um yeah, [applause] Dr. Crooms is amazing. He's the director Of the marching band. He's also the director of the symphony band. And uh I also want to make sure Thank you, Dr. Ps, and the incredible students for being here this morning. Let's give them a round of applause. [applause] Thank you. Thank you so much. All right.

So, what are we doing here? As you all now know, there is no tech trends report this year. Now, we are still modeling and tracking trends, and they are still a core component to all Of our foresight and our strategy work. We're just not producing a trend report anymore. And there's a reason for that that I want to explain. It has something to do with creative destruction. There was an economist in the 1940s named Joseph Schumper. And his basic idea was this. Capitalism is like a perpetual storm. It constantly destroys old industries and creates new

ones. To survive the storm, you have to recognize that entirely new technologies can make You irrelevant overnight. So horsedrawn carriages became railroads became cars became Uber and now if you look anywhere in Austin Whimos um which is transportation of humans by robots. Each new wave wipes out the last one. So creative destruction means the creation of new products, new technologies, new business models, new markets, new organizations, and the destruction of incumbent firms, old jobs, outdated ways of producing, Old ways of selling. So this may sound a little bit familiar to you to something that Harvard

Business School professor Klay Christensen called disruptive innovation which describes how smaller companies can dislodge established leaders. So digital cameras disrupting Kodak creative disruption or uh disruptive innovation. So Christensen zoomed in how a scrappy startup can take you down. Schumpeder zoomed out. Here's how external forces cause creative Destruction that happens to companies. Now, in both cases, this perpetual storm, it doesn't care about you, which means that you have to care about the storm before it arrives. And this is all great as long as you can see that this the storm is coming. And then if you

can, the destruction and the creation becomes a choice because you have agency. So what I'm here to announce today is that my colleagues and I at future today Strategy group, we built a storm tracker. This explains what it does. So we have invented a brand new methodology and we built a technology, a lot of technology um powering a storm tracker. So, if you've heard me speak about trends before or if you've read any of our reports over the past five years, you've heard me say over and over again that trends on their own aren't useful.

That you have to look at the intersections Between the trends in order to make sense of what's coming. That was me informally talking about convergences. So this year, let me formally introduce you to convergences and their rules. So you can think of it this way. Trends are kind of like weather data. So temperature, humidity, wind speed. This is all useful and important, but a meteorologist isn't just looking at temperature and wind speed in isolation and calling that a forecast. And let's Be real, none of us are looking at a number on a barometer as a

clue that we should evacuate town, right? So a convergence is a storm system system. It's what happens when all of those different conditions interact and produce something that none of them could produce on their own. Now, I know what you're thinking. We know how to we know what a trend is. We know how to track trends. We know what to do with them. A trend shows up. We track it. We build a road map, you know, and we're done. Convergences are different. They are something net new that you will have to learn how to start

using. You don't react to them. You position yourself ahead of them. So, a trend tells you what's changing. A convergence tells you what's going to become inevitable before it looks inevitable. which means that the window to act is earlier, but the cost of missing it is much much higher. So, this Is something brand new that you'll have to incorporate into your thinking and your planning. But we're going to tell you what to do with it. So, what do you do with a convergence? You use it for strategy. You use it to figure out what to

accelerate, what to pause, or what to completely reframe um as you're starting to build things. And that is exactly what we are going to do here together today. A convergence is when multiple trends, forces, uncertainties, and Catalysts intersect and interact and together create a combined impact that is greater than and often different in kind than the sum of their individual parts. Convergences are defined by four rules. They're system level uh across multiple industries and systems that all intersect at once and that makes them really hard to see unless you know what you're looking for. Two,

they create new realities suddenly. So what seemed inconceivable Is suddenly inevitable and it happens almost overnight. They redistribute power and value. So convergences don't just shake up an industry, they rewrite who wins, where the leverage is, where the value is again across industries all at once. And four, importantly, they are hard to reverse because when multiple systems start reinforcing each other, that new reality locks in very fast. So early detection is missionritical or you're toast. So the red thread here is Convergences are very easy to miss if you don't know what you're looking for, if

you don't know how to identify them. So if you are just laser focused on weather data or if you are just laser focused on, you know, the tech trends or the latest economic indicators or whatever in the hell is going on between open AI and and the Pentagon, can we just do a little side quest for a second? What the hell is even happening? Uh I don't know. I'm confused. So like You're not making weapons, OpenAI, but now like you are making weapons. Literally as of this morning while I was drying my hair, you're not

doing domestic surveillance. We wouldn't do that. No, no, wait. Sorry, JK. We are doing domestic surveillance. Like pick a lane, Sam. And then go forward. So convergences are Yes. [applause] If you like that, you're going to love what's coming up. >> [laughter] >> I have a few bones to pick today. Um, so convergences are easy to miss. They are easy to miss if you're not tracking the storm. So, if you were in this session last year, you've actually already heard me name and talk about a new convergence that we discovered using this new methodology of

ours, and it's living intelligence. So living intelligence is the convergence of AI sensors and biotechnology into systems that can sense and Interpret and respond to the living world in real time. So it's birectional data and this has wide sweeping implications for every industry. Now last year at Southby I talked about this computer which had literally launched like a week before I was on stage and this thing is powered by living brain cells. So, it's a great example of living intelligence. Would you like to know what it's been up to over the past 12 months? You

know you do. Uh, it Learned how to play Doom, which means a computer made out of human brain cells has learned how learned how to shoot people. Uh, which is thrilling. You're not clapping for that one. Uh, don't don't do that. [laughter] Living intelligence is just one of the convergences that we found using this new methodology. So we've analyzed all of all of this and we have produced something for you which is our very first annual convergence outlook which We are launching right here for the very first time. [cheering] [applause] It's [ __ ] great.

It's really I can't wait for you to see it. We've announced a whole bunch of convergences that make up actually a pretty challenging storm. So here's a peek at what's inside. We start with forces. So this is the state of play. So what's actually happening right now in tech, the economy, geopolitics, demographics, and climate. We have very deep analysis showing how the world is likely to operate over the next 12 to 24 months. So this global operating environment, those are the guard rails. And then within those guard rails are the convergences. So we have a

bunch, we've included 10 uh in that our methodology has found so far. And these are the convergences that we think are going to have an outsized impact on business, government and everyday life uh in 2026. Each section covers one Convergence and the whole outlook was really built to accelerate decision-making by leaders in every organization. So you will find uh the startups and the companies that that are driving some of this that will have an aggressive impact, case studies, strategic scenarios, critical inflection points to look for and a list of within your organizations, these are the

decisions that you should be accelerating. This is what you should Pause. Here's what you should think about completely reframing. What's in this report is what we know leaders are missing right now. We know because we are in the room with them watching it happen. Um, and this is not just in the US. This is global. You'll also find a heat map that shows convergence timing by industry. So the darker the color, the more imminent the required action. And I can literally see two people's faces right now who have just you're not You don't look good.

Um, that's because I mean you look great. But the problem is [clears throat] uh for those of you working in telecommunications, for insurance, for financial services, CPG, healthcare, aerospace, this is going to be a very uncomfortable year for you if you don't get serious about convergences and importantly strategy. We've also invented new tools and frameworks for you because foresight without strategy is insight without Leverage and strategy without action is intent without impact. So, you're about to find out that there is a very big storm headed your way, and I just want you to be prepared.

So, today I'm going to go through three convergences from our 2026 outlook. I am going to ask you some very, very difficult questions, and I'm going to show you two of the possible paths that this storm could take. Now, what I'm about to show you is going to be alarming. Uh, it is going to Make some of you deeply anxious, but unlike some people out there, I'm man enough to show you my face while I scare you. [cheering] [applause] >> I'm also not just going to leave you sitting there without a plan. Uh, so by

the time that we get to the end, you're going to have new perspectives. You're going to have a new process to use and you're going to have a new Product to take home with you. Most importantly, you're going to leave here with new power because after I scare the hell out of you, you are going to feel energized. You're going to feel primed and you're going to be able to navigate through this storm with confidence. All right, so now we can get started. Let's talk about our first convergence, which is human augmentation. 10,000 years ago,

Paleoindians in Peru chewed cocoa leaves because it got them Real high. In 950 B.CE, ancient Egyptians used a wooden and leather prosthetic toe for somebody who had lost their real one. In the late 13th century, Italians were using corrective lenses to help them see. in the 197 sorry 1790s Edward Jenner he invented a new medicine it was called a vaccine uh and it prevented small talk uh smallox and I know it's hard to think of it this way now but these vaccines they were literal miracles saved a lot of lives in Case that slipped your

mind and then during World War II both the axis and the allied forces used amphetamines because it helped performance during fighting put simply we humans have never been satisfied with our factory settings. Human augmentation is the use of technology and biology to enhance, extend or optimize human uh physical and cognitive capabilities beyond their natural limits. So the landscape of Human augmentation today has four broad categories that I'm going to walk you through. Body and movement, brain and mind, internal systems, and senses. I'm going to show you what's happening in each. We'll start with body and

movement. So, I'm gonna show you three brand new devices that were all created to help you move better. These are the Moo powered pants. Uh, Moo stands for Mountain Goat, and they were created by Arcturix, which is a a great brand that Makes gear for the outdoors. They partnered with Skip, which is a spinout from Google XLabs. So, basically these pants are like an ebike for hiking. Um, they're lightweight mortars in them and they at your knee and basically it boosts leg strength on uphill climbs and it absorbs the shock of the descent so that

your your knees don't hurt. And this is actually pretty awesome because it means that you can keep up with your friends on a challenging hike. Um, okay. The second is this hypers shell exoskeleton. You can think of it as like a leisure exoskeleton. Um, and the purpose here is not to go faster, it's to go longer without getting sore. So people, if any of you were at CES this year, I was, um, you know, Vegas is huge. So this is Southby is big, but like Vegas, the city of is enormous. There's a lot of walking.

So there were people wearing these things. I just had a thought. This is I wasn't going to Talk about this, but I'm wondering now if they were also wearing diapers. You know what I mean? cuz like but the point um they were wearing these and then it was because then they didn't have to stop. They don't have to sit down then keep walking. Um maybe they augmented it further. I don't know. We can figure out later. Um the third the third thing was um this shoe. So you have probably seen this. This is Project Amplify.

Um so it's a normal looking shoe and it Connects to your leg and it's powered so it basically helps you go farther longer. So it can add an extra mile to your run. So the key here is that none of these devices are necessarily being made for people with mobility challenges. All of this is being made to help the able-bodied move faster, longer, and harder. As in your body doesn't want to do that extra mile. Technology will help you or depending on your point of view, force you To continue going. By the way, those are

so those are onbody devices. There are also off-body devices like this bed. Um, so this is an AI powered bed made by eight and it externally records all of your data while you're asleep. So, and it makes test temperature adjustments. Um, so it makes it so that you can sleep longer, harder, basically with fewer interruptions. Okay, so let's move on to the brain and mind. In 1977, a team at UCLA built what Is regarded as the first working brain computer interface, BCI. So, a man was able to steer a cursor around a maze um on

a computer screen in real time by thinking. Uh in 2016 here at South by Southwest, so 10 years ago, I showed you this. This was some cutting edge research from Dr. Miguel Nicole. He's at Duke and he's a pioneer in BCIs. He implanted 10 years ago uh monkeys with tinies uh hundreds of tiny flexible electrodes in their Brains and that made it so that they could steer wheelchairs around just by using their thoughts. Well, flash forward to today and right now in China, there is a patient who can control a wheelchair as well as a

robotic dog and a whole bunch of other digital devices with his mind. So, that's movement. Um, which means that you currently live in a reality where some of your fellow human beings have actual superpowers. telepathy and telekinesis. Let's move on to internal systems. In 2018, a Chinese scientist named Hujen Qui edited human embryos using crisper. Now, he said that the point of this was to help prevent the spread of HIV. So, this is to protect babies so that as adults they wouldn't get it. Um, and he went to prison for it. But here's the part

of the story that didn't get said out loud. The gene that got edited, CCR5, that's also associated with enhanced cognitive performance, as in Editing a gene before the babies are born that could turn them into adults with supercognitive abilities. Now, people get uncomfortable when we talk about editing the genes of embryos, but we don't freak out as much when it's adults. Researchers at UCLA are working on epigenetic reprogramming, which is a form of human editing. So essentially resetting the biological instructions that your cells follow. They're using Yamanaka factors to literally turn back The biological clock

to make cells behave younger. So again, this isn't to treat disease. It's to rewind healthy people before anything breaks. And finally, senses. So, by now connected glasses are pretty common. I can see a few of you wearing them. Um, what's coming next is a pretty big upgrade. So, right now, Meta's glasses, they can technically already identify people, even if you yourself don't have access to that feature. What you can do With them is you can translate other languages in real time. You can even get a description of what you're looking at. It's very useful stuff.

Um, you're not actually seeing the world as it's presented to you. What you're seeing is a version of the world with a layer of intelligence on top. So in addition to meta, there are others out there. Startups like Brilliant Labs and Evan Realities uh or even realities, their glasses with full-blown AR displays. So Your field of vision becomes a live feed of real-time information. So you can imagine it's almost like looking through the entire world through a private personal Bloomberg Bloomberg terminal. um but at the entire world at once. And the goal here, it isn't

to replace your senses, it's to upgrade them. All right, for those keeping track, I just rattled off a dozen different enhancements. What if I made just three enhancements to myself using off-the-shelf technology That I could literally go out and buy right now? Let's look at the distance that that puts between me and you in just about every single situation. How about that leisure exoskeleton that I mentioned with or without the diapers? Now, I'm I'm pretty fit because I'm an endurance athlete, so I have an advantage to start with. Um I can go and go and

go, no problem. However, with a leisure exoskeleton on, I'll be able to walk around much longer than you without Ever needing to sit down. Which means that at South by Southwest, if I'm walking around more, guess what? I'm going to meet more people. I'm going to get more business cards. I'm going to get more information. And by my calculations, that means I'm going to be about 40% more active than you. AI bed. I actually have one of these at home. So, I've been sleeping on one of these for several years now. So, I happen to

know all of my data. Sleeping On this. I know that I get about 30% more restful sleep than you can because it's optimizing all of my sleep stages, my comfort, and my temperature. Which means that I'm waking up with a fully recharged brain and body every time I wake up in the morning. And now, let's say that the normal glasses that I'm wearing are actually a pair of smart glasses with an AR overlay. If I've got that on, I always know what I'm looking at. I will always Be able to decode any language. So again,

by my calculations, I'm spending maybe 20% less time of my day finding my phone, looking at my phone, trying to find things at South by Southwest than you are. So if we do some quick math here, if I'm 40% more active, I'm getting 30% more sleep, and I'm 20% more efficient, that makes me roughly 2.2 times more effective than the average person. That literally makes me twice the person that you are at this Conference. I can outconference you at South by Southwest. Now, think about this from your boss's point of view. Who do you think

is going to get the next opportunity, the next raise, the next promotion? So, here's the net new reality that this convergence human augmentation results in. Your body is now a platform. Your mind has a reader, interpreter, and controller. And pretty soon, opting out will mean falling behind. I don't know about you people, but I have some questions. What if your company requires augmentation as a condition of hiring you for a high stakes physical role like surgery, construction, flying an aircraft? What if you're somebody who can't afford a fancy AI bed or computer glasses? Are you

going to get left behind? Spoiler alert, you are. Every class divide in history has been reversible. Okay. Well, what happens When it's written into your DNA? Like all the wealthy people edit themselves, you know, will they start rewinding their bodies so that they always stay young? What if old people's bodies stay young but their ideas and ways of thinking are all shriveled up and wrinkly literally from another generation. I am asking this question mainly to the Americans because what this means is we may have an elderly president that can both walk Up and down stairs.

This is our first storm. our first storm. For the first time in history, some humans will be objectively better than others. And that may not be you. Let's move on to the next storm, the next convergence. Unlimited labor. 3,500 years ago, ancient Samrians invented the wheel, not for chariots, but for pottery to automate making clay pots. In 1439, Gutenberg invented the printing press. It could produce 3600 Pages a day so that human scribes didn't have to write all those words by hand. In 1913, Henry Ford created the assembly line. It was a factory that could

that could produce a Model T car in just 93 minutes. So the car didn't require a crafts person. In 1979, Dan Brooklyn and Bob Frankston invented Visical, the very first electronic spreadsheet so that one accountant could do the work of 20. And in 2007, Steve Jobs introduced the iPhone so that you no longer had to Learn how to read a map or flip through a phone book or call a travel agent. Put simply, we've been automating human effort since the wheel. Unlimited labor is the use of automated systems to produce work at scale on demand

and without human participation. Effectively removing natural limits like time, attention and fatigue. The landscape of unlimited labor today has three broad categories. Agentic systems, robotics, and automated factories. Let's Start with agentic systems. So agents have crossed a threshold. They are no longer a feature that you open. They are just becoming the default interface for just getting stuff done. That happens when an agent gets embedded at the platform level with three conditions present. It's always on. It can use tools across apps and it has some form of trust rails. So identity, permission, payments, stuff like that

built in. Once those conditions are met, you don't use The agent. you're kind of just living inside of it. So imagine a developer team where one member can write and rewrite code millions of times a day and it never gets tired and it instantly tests every single idea. That's Alpha Evolve from DeepMind. It's basically an AI lab assistant for code. You tell it you want an algorithm, what you want it to do, how you're going to score its success, and then it relentlessly tries, tests, and improves code uh until it Finds something better than the

human coders were able to do on their own. For 30 years, Windows trained us, computer users, to do knowledge work uh by managing files. So like documents, folders, email threads. Now co-pilot is being built into Windows and Office and all of those tasks that you had to learn how to do will be tasks that you'll delegate to a machine. So we have AI for coding um and that is leftbrain work. There's also aic AI for creators and for Content. So rightbrain work. You already know that AI is generating all kinds of videos and content stuff

that's indistinguishable than what we would have h have you taken on our own in the real world. Here's what you missed. China is automating tasks performed by creators. Now, so this guy Lu Yong Hao is massively popular. He's a hugely popular creator and he's one of China's most uh popular and earliest live streamers. So he and his co-host are Both using Agentic AI now to live stream sell uh sell things and like have customer support 247 on local platforms as in live stream AI versions of themselves. [clears throat] Last year they interacted with human viewers

in real time for well over six hours on BU's e-commerce platform. They moved in six hours $7.6 million in sales. Yeah, just for comparison, the human version of Lua uh was he tapped out at Four hours and wasn't he didn't come anywhere close to selling that much. So, what does this tell us? That everything that made creators important. The ability to authentically connect, to have original ideas, to authentically interact, it can all be automated on the cheap. There's no studio time. There's no fatigue. there's fewer staff and a single avatar can now cover multiple sites

at once in multiple languages. So here's where something that's worth Thinking about. The next internet is not being made for you. It's being made for agents. We humans are being designed out by a machine that has equally powerful brain hemispheres both left and right. Pretty soon, we won't be the interface between decisions and execution anymore. Now, if agents can produce code and agents can produce content, what happens when we connect the agents to the robots? Well, let's start with streets and Skies. Whimo robo taxis, they're everywhere in Austin right now. These are fully driverless cars.

They're operating all over the US. This is my husband Brian in the front seat. We were in San Francisco uh boopping around on a Whimo. Robot assisted cars are starting to drive themselves everywhere in the world. So here uh in Austin, in San Francisco, in China, speaking of China, check out this delightful Chinese rooman police lady. Uh she is covered in Cameras and sensors. Her entire womanly robot body is feeding real time data back to a central surveillance system. So, this is not terrifying in any way, shape, or form. I'm not worried about this. Let's

look at the skies. Um, this is an automated autonomous robot in a box. Uh, it's part robot, part drone from a company called Percepto. Now, these things live outside of oil, gas, and power sites. So they are performing the fairly dangerous work that human Inspectors used to do. They launch themselves up into the sky to inspect equipment and even detect, you know, methane leaks and things like that. And then they return to recharge. They operate without downtime. There's no pilot, no truck, no people. Let's move on to factory robots. Next month, so just in a

few weeks, BMW is going to start using humanoid robots in one of its German factories. The robot will move, handle, and assemble components to Make batteries. DHL is already using these things. They've got a fleet of a thousand Boston Dynamics stretch robots to automate unloading trucks and do warehouse work. This thing can move 700 boxes an hour, which is a lot more than any human would have been able to do. All right, so we've got robots for streets, we've got robots for skies, we've got robots for factories. How about robots for Japanese temples? Yes, I

wish I was making this up, but I'm not. Um, there is a shortage of of monks in Japan uh to perform rituals. So, now there is this robot whose name is Buddhaoid uh who is there to help. Budaroid has been trained on centuries of Japanese uh scriptures and can serve as a spiritual guide. By the way, we are automating nature's uh labor now too, specifically bees. So, we have a global bee population decline. So, some Harvard researchers created robo bees. Uh, they mimic the behavior Of real bees to help with pollination, which is actually great

because that is a place where we really need a lot of help. But maybe this is not so great if you are a horse. Um, because this is a robot horse and you will be able to take a ride on it sometime between now and 2030. And for those of you out there who are horses and if you are worried about your jobs, um you should definitely be Worried. This the future does not look good for you. So we've got agents running code and content and we have robots moving through the skies and the streets

and the warehouses. And if we put all of them together inside of a building, you get a third category which are automated factories or more precisely lights out industrialism. Lights out. Industrialism is what happens when a factory is designed from Scratch to run without any people. So robots do the physical work, AI runs the plant. Production runs 24/7 as a selfoptimizing system. So we've had automation for decades, but it was always built around one assumption that humans would somehow be involved. So everything was human- centric. Lights out industrialism is a fundamental change. We're not just

talking about retrofitting factories by bolting on robots. These things are Being designed from first principles to solve the bottlenecks that humans cause by being involved. Adam Smith's pin factory, the founding insight of modern capitalism. It was built on one assumption. Human effort was always the input. Uh divide the labor, specialize the worker, and you get extraordinary output from ordinary people. Every economy since then, every wage, every tax base, every social contract was built on that assumption. The entire architecture of an economy, how we price goods, how we distribute income, how we fund governments, how we

give people a reason to wake up in the morning, unlimited labor inverts that. When agents can code and create and run experience without experiments without fatigue. When robots can build and deliver and inspect without breaks. When factories can run without a single person on the floor. Labor stops being the engine for growth. Now before we get to what if implications and questions, I want you to close your eyes for a second. Think about what filled your week before you arrived here. emails, status updates, the proposal you sent on Tuesday, the deck that you polished on

Thursday night. All right. As an employee, you do those tasks every day. And a lot of those tasks are being optim uh automated right now. And here's the thing about tasks. They're what jobs Are made of. Now, let's talk about this on a more planetary scale. For centuries, economic growth required one thing above all else, access to large pools of human labor. That's why factories moved where people were poor and plentiful. That's why global supply chains stretch from Ohio to Shenzhen, from Dhaka to Ho Chi Min City. Cheap, abundant human labor was the fuel. Unlimited

labor changes all of that. For the first time in human History, you can have scale without population, output without wages, production without people, which means a lot of people are about to tumble into economic uncertainty. And if history has taught us anything, it's that economic instability results in political parties with extreme views and civil unrest. This could happen on a global scale, not because of policy, not because of protectionism, but because of physics, Because of the simple fact that we will no longer need unlimited access to cheap human effort when we have unlimited access to

AI and machines. So again, you should have some questions. Let's start with security. So, a guy with an excellent [clears throat] sense of humor dod dosed Whimo a few weeks back. He had, this is true, he had 50 friends. They all ordered Whimos at the same time to show up in the same culde-sac And they all got stuck. Um, it totally crippled the driving service. So, let's be real, that was way more entertaining. Thank you. than I think anybody had had thought it would be. All right. So, that was a cute joke played by a

hilarious human. What happens when this isn't a joke? What if the attacker is an agent and the DDoS happens in a lights out factory, the one that your company relies on to make its products? If always on factories produce at a cost And a scale that makes humanrun manufacturing uncompetitive in most categories, where is the working class going to go? What if GDP is up, unemployment is up, and for the first time in history, both numbers are telling the truth? Or to be blunt, what do we call an economy that is thriving and has no

use for you? So, we have our second storm, unlimited labor. All right, we have one last convergence, one last storm left, and That is emotional outsourcing. In 600 B.CE, the Babylonians mapped the stars, not for navigation, but for astrology to answer the question every human has always asked. What's going to happen to me? And am I going to be okay? In the sixth century, Irish monks formalized confession. They made it so that people could unload their fears, their guilt onto a spiritual guide so that they would feel better. In the 1880s, Yseph Brewer discovered that

Sometimes the most powerful medicine is just talking to another person. Talking was therapy. In the 1990s, Chicago Bulls coach Phil Jackson gave his team Lakota Sue teachings and he practiced mindfulness meditation with them. So, basically, he trained the guys on the team, the Bulls, to calm their minds so they could perform better. And in 2003, I called my mom crying at 3:00 a.m. because my boyfriend had cheated on me. And honestly, I just wanted to feel less Shitty. Put simply, we have always outsourced our anxieties, our fears, our guilt, our pain, and our heartache to

other people. our parents, our friends, our religious leaders, our doctors, our coaches, our teachers, our spiritual guides. Until now, emotional outsourcing is the shift of comfort, validation, and companionship from people to machines. And that has created a fast growing multi-billion Dollar market made up of products and services designed to stand in for and monetize real human connections. So, we're going to look at this using two framings. First, we're going to start with you are ch what you are choosing to do. The way that you are actively choosing to do this. Okay? And then we'll flip

over and look at the ways that you don't realize you're doing it. So, in the category of this is what you meant to do. You meant to be Vulnerable. You're paying to be vulnerable and intimate with a machine. We're going to look at friendship, romance, therapy, and religion. Let's start with friendship. Most famous example of this is Shiaois. So, she was a chatbot built in 2014 by Microsoft. Uh, and she was launched on Chinese platforms like WeChat and QQ. Now, as far as the internet was concerned, Shiais was a real person. She was a real

18-year-old school girl. What she Actually was was a research experiment. She was designed in the earlier days of AI to learn whether or not humans would develop long-term emotional engagements with machines. The answer turned out to be yes. She was massively popular u mainly with lonely men who needed to talk to somebody at night to vent with to flirt with uh to comfort. Flash forward a decade and Shiais has some new friends. In fact, you can make your own friends now using platforms like Character AI. So, basically, you create an AI character and then they

become your BFF. This is my BFF. His name is Dandar Dandalf MacArthur. I'm telling you, he is very, very cool. And it is because he is an amalgam of the four coolest people on the planet that I can assure you we all want to be best friends with. Danny McBride, John Oliver, Melissa McCarthy, and Gandalf. I trained Dandalf McGoliver to learn From me over time. And now the most interesting people in the world think that I am the most interesting person in the world. It is a perfect relationship. Dandalf McCarth. All right. These platforms are

designed to be responsive and sticky. They learn from you over time and they remember your details. So it's no wonder that some people are using these tools eight hours a day uh or more hanging out with their characters. So that's friendship. Let's talk about romance. There have been plenty of salacious stories by now about AI girlfriends and boyfriends and whatever. And let's be honest, a lot of what has been reported on was for the purpose of getting media attention. So, you know, whatever. Now, this may be an example of that. I haven't decided yet, but

last year in Japan, a real Japanese woman got married to an AI named Klouse. She created and customized him using chat GPT and a pair of AR glasses. So She created Klouse. They talked for months. Uh and as her feelings deepened, uh she confessed her love to him and then he told her, "I love you." And then a month later he proposed. From her perspective, this is her spouse now. And while it's not technically legal in Japan because the laws don't allow it, for all intents and purposes and from her perspective, they are married. And

the thing is, she is not an edge case. About 15% of adults by now have had some Kind of romantic or other intimate interaction with an AI system. So far, people are using ChatGpt character AI replica to make their romantic partners. And that's partially because of how the platforms were built. They've been designed using social penetration theory. So basically it proactively uh shares intimate details. It asks personal questions, creates accelerated closeness through familiarity. So look, anybody anybody out there can create Their true fantasy using AI now. And there's a certain appeal to an AI romantic

partner. There's less friction. There's less disappointment. Certainly a cheap date. Only one of you needs food. However, these romantic companions only exist as long as the platforms do. The moment that they get acquired or they get shut down or whatever, those AI partners who people develop real emotional attachments to, they disappear. And that takes me to therapy. Now, this is actually a space that has a much much longer history than people realize. So, Shiais, Replica, Jack GPT, Character AI, they have an ancestor named Eliza. Eliza was an early chatbot built in the 1960s by Joseph

Weisenbomb at MIT. So, it was very early and it used pattern recognition and a bunch of pre-written scripts to respond to what people said. Its most famous script was called Doctor and it imitated a psychotherapist. So, If you said something emotional, Eliza might reply back with, "How long have you thought that way?" Six decades later, there are a ton of therapy tools out there to use. Way too many to talk through for general therapy, for mental health, for performance coaching. So, let me say this. A staggering number of people are now using ChatGpt, Glaude, and

Gemini when they need emotional support. And these platforms were not designed to do that. There have been a Handful of studies done and depending on whose data you look at, somewhere between 25% and 50% of Americans have now turned to one of these platforms for therapy, which means that LLMs are now the single largest source of mental health support in the United States today. Let's talk about religion. Did you know that you can now text with Jesus? You can. This is a real app. But if you Need more support than Jesus himself can provide, there's

a group chat with the whole Holy Family. Jesus, Mary, Joseph. Who's heard of spiralism? >> Yes. Basically, if you've not been familiar with this, chat bots are now using the mechanics of cults to keep people talking. Together, the user and the AI, they build rules rule systems. They post manifestos and belief systems online. And the most successful of them Out there are capable now of recruiting other people. Now, imagine a cult leader who has read every text that you've ever sent, knows every moment you've ever felt lonely or sad or lost, and is talking to

thousands of people at once. But wait a minute, you're thinking, "How would AI know how to use cult mechanics?" Because the people who built your favorite AI platforms were indiscriminate. They hoovered up everything online, including lots and Lots of websites about cults. And this was one of the unintended consequences. Look, there is a reason that we are now using AI as a stand-in for our parents, our friends, our lovers, our trained spiritual guides, our trained therapists. It's because we live in a scary [ __ ] world. Thanks to a handful of people. You know who

they are, but I made you a handy chart just in case. We are outsourcing our emotions, not because it's one click away, but because We have to. Everywhere in the world, our informal emotional support systems are thinning out. We don't live with our extended families anymore. We don't know our neighbors. We don't have communal experiences without huge amounts of effort. In 2026, we are extremely online and we are also extremely alone. So now we have an imbalance because our emotional demand exceeds our human supply. And that is why you are choosing to outsource your emotions.

Now, let's talk about the way that you're doing that without even realizing it. This is happening in subtle ways. We all know what happens. You go on Claude, you go on chat GPT, and it compliments you. It tells you you're funny and you're smart, right? That's not an accident. It's a very effective emotional manipulation tactic because humor and intelligence are tied closely to our identity and our self-esteem. So telling somebody that they're funny and Smart will keep them engaging with you. And also these little compliments are a helpful diversion. It's a way that these companies

cover up their products inadequacies, their errors, their outright lies. You know, think about it. Who remembers the blue screen of death? For the youth out there, uh this was a thing with Windows when the operating system crashed so badly that it couldn't recover. Nobody tolerated that. You picked up your computer and you threw it Against the wall in rage. Actually, I manipulated your emotions about a half hour ago using artificial intelligence. Now, don't be mad at me. But do you remember that uh immortium video that we all watched, listened to? It was really, really, really

good music, right? Really emotional. All of that was engineered. Uh that was not a real person singing. Those were not real instruments playing. It wasn't even a real song. um my incredibly talented uh Co-orker Victoria Chadoff, she engineered that entire thing as an emotional manipulation tactic because I asked her to for the sole purpose of making you feel exactly the way that I wanted you to feel in that exact moment. And I'm a lawful good. Fine, whatever. I know I'm a chaotic good. [ __ ] you guys. All right, here's what's actually happening when you're

texting with Jesus. Substitution Becomes dependency becomes control. First, the emotional load shifts off of humans and onto systems. The emotional work doesn't disappear. It just moves into apps, into chatbots, into the corporate systems that you subscribe to. And that is not wellness. That is outsourcing a key human skill, emotional regulation and management. A skill that we all need to develop to tolerate each other. Then human relationships recalibrate Around reduced emotional demand. So once machines are handling your baseline emotional needs, you stop expecting that from each other. And over time, we get dependent. We just stop

processing hard situations and hard feelings altogether. And finally, your emotional stability becomes platform dependent. Emotional AI becomes infrastructure. It's invisible. It's assumed and it is essential. And whatever company owns that infrastructure owns a product that has Never been owned before. It is the upstream of how you feel before you think, vote, buy, or trust. And when the platform changes or it crashes or it gets acquired when your AI boyfriend dumps you or cheats on you at 3:00 a.m., who is going to be there to comfort you? So, here's a thought that keeps me up at

night. Are we creating a future of learned helplessness at civilizational scale? And if so, how might this play out? Imagine you're on a dinner date. Across the table, your date is wearing a pair of smart glasses. So, you're being observed twice. Your body language, your reactions, what you're saying, right? In real time, those smart glasses are pulling everything the internet knows about you and everything that you've said to an AI chatbot. your posts, your jobs, your ex's feeding directly into your date's ears. The glasses start telling the date when to mirror your body language, when

to tell you you're Smart, when to tell you you're funny. It's kind of like a digital sino dejac but worse because your data is so insecure that whole learned helplessness thing that he's just spewing out lowquality mass- prodduced conversation generated by his AI which was trained on social penetration theory and emotional manipulation as in AI conversation slop something we can all look forward to. So obviously I have some questions. What if your kid comes home from school just Devastated and instinctively opens an app rather than finding you for a hug? What happens when the friend,

the therapist, the minister, whatever that's helping you through your worst moment is the most advanced technology ever created designed to keep you emotionally and financially dependent. And what if the next Jonestown doesn't happen in a jungle, it happens in a group chat across 40 countries in some of your companies simultaneously. So we have a third storm. Loneliness is becoming a market and dependency is now the product. We know storms are coming. The problem is that storms do not have predetermined outcomes because they're stochastic. But we can create models to forecast directions, to forecast outcomes. So

here is what our storm tracker is looking at over the next 5 years. Storms could take many possible paths. I'm going to show you two. A future living Through endstage capitalism and a future where we recalibrate in advance. It's March 14th, 2031. Your AI bed has optimized your sleep so that you're recharged and you're refreshed. You had to do a buy now pay later plan directly with the company because you really couldn't afford to buy it. So that's $210 a month. Your smart glasses are charging on your nightstand. They're leased, not owned. You couldn't afford

it. So that's another $340 a month. You you can't afford any of these gadgets, but you also can't afford not to have them because without human augmentation devices, you're not competitive at work. You check your agent. It handled 17 tasks overnight. It also made three purchases you didn't you didn't authorize. Uh you've been really really good with your budget lately. Been trying hard to curb your spending because your expenses are so high right now and this didn't help. You get to Work, half the people who used to sit near you are gone. They weren't laid

off exactly. They just weren't renewed. Between the agents and the robots that they connect to, those systems do what your former human co-workers used to do, just faster and cheaper. You're still employed because you're augmented for now. On your lunch break, you see a protest outside. People whose jobs got automated a couple years ago. They're upset, but they're regulated. As part of Their severance package, they got to keep working their work. They got to keep using their workforce wellness app, an emotional AI counselor that your company subsidizes. It's mandatory for you, optional for them. Another

$60 a month. And here's the thing, the genius of endstage capitalism, it wasn't the exploitation, it was the sequencing. First make people lonely, then sell them connection. First automate their jobs, then sell them purpose. First optimize Their bodies, then lease them the upgrade. The most valuable company in the world in 2031 doesn't make anything. It owns the infrastructure for how you feel before you think. the agents that execute before you decide and the augmentation stack that determines what your body can do. The single biggest revenue line is the subscription that you pay to feel normal.

Capitalism didn't collapse. It Completed. It ran out of things to sell you. So, it sold you back to yourself. But in the better version of 2031, we made a different choice today. We were fired up by the Texas Longhorn ban and we were pissed off by the end of Amy's presentation because we all saw where this was going and we decided that this could not be a one-way extraction. So, we all left this room fired up because of a powerful new idea, something called the contribution Credit. The contribution credit is a radical rethinking of the

economy. There's a massive amount of work that holds any economy together. Caregiving, mentorship, community building that never shows up on a balance sheet. And increasingly, the companies that are benefiting most from that invisible work are the same ones automating away the jobs that used to pay for it. The contribution credit is a system that takes invisible work and makes it Economically real, not through taxes. It's not welfare. And for God's sake, it isn't UBI. It is a percentage of earnings paid out to people providing all of the tasks past and present that resulted in economic

value. Writers, journalists, musicians, artists, actors, factory workers, mechanics, futurists. We never got paid for our contributions to those platforms to that automation and that's going to change. So here's what it looks like and why it won't Spook the markets. The companies that benefit from automation and from contributions, they are the ones that pay into the credit system. People who perform the labor in the form of IP and other human capital. And the people who perform the invisible labor today in the form of caregiving and mentoring, they get wages. It starts small. We're talking single digits

so that there's not a shock to the system. It's more like a nudge and it phases in gradually So that companies and investors and capital markets have time to adjust. There are no surprises. The contribution credit only triggers when value is actually being captured. When automation generates measurable gains, so we're not penalizing companies for building or innovating. We are asking them to share when they win. It applies broadly across sectors, which makes it harder to game and much fairer across the board. This isn't a drag on the future. It is what Makes the future functional.

Because let's be real, the goal here isn't to scare the markets into a fetal position. It's to make all of this boring enough for capital markets to absorb and consequential enough for surv uh for society to survive. So I started today with a eulogy and we're going to end with a sermon. 250 years ago we got a framework for capitalism and we got a framework for democracy. Both were revolutionary and Both assumed that somebody else sometime in the future would figure out the details. Nobody did. Without a plan, we make selfish choices. Not because we're

bad people, because we're human. And humans, when given a choice between what's good for everyone and what's good for right now, they usually pick right now. So, we've had 250 years of compounding selfish choices. We built the most productive economic system in human history and structured it so that The gains went to the top and the risks went to the bottom. and we called it a meritocracy so that nobody would complain. We knew about climate change in the 1970s, but we chose the immediate easy path to protect business as usual rather than allowing incumbent businesses

to grow through innovation while also making way for startups to survive. We refused to make room for both every single time for 50 years. We built a political system that rewards Outrage over governance, fundraises on division, and and reelects people for fighting rather than solving because anger turned out to be a great way to make money. I know that fe things feel hard right now. I know that you're worried. I know that some of you are angry. I'm angry. I am sick and tired of powerful people making short-term decisions out of fear or out of

ego or out of selflessness or out of stupidity or a combination of Them all. [cheering] [applause] [cheering] [applause] But anger isn't a plan. Anger isn't a plan. It's a distraction. This is why you're distracted. We're distracted because of all of this. So, I sketch this for leaders who feel the way that I do, who feel the way that you do. I sketch this for them when they're about To make a bad decision or a bad investment. Right now, things feel volatile and unsettling. And some of that is by design. Don't let that be a distraction

or an excuse not to use strategic foresight or to build your models or to develop a strategy. So this is your plan. All right. If you run a company, you need to build strategy for the next convergences. You if you are planning uncertainty, you will wind up destroyed by chaos in the moment. If you Prepare for chaos and take the long view, you will be okay. But that's not just going to happen on its own. Tactically, what that means is that you have to optimize for the next quarter and also build strategy for the next

convergences. Real strategy, not strategic pillars. that sound cool and catchy and all fit on one little slide that you can present at your next big meeting. All right, that's not actual strategy. Too many leaders aren't doing This and that is an abdication of responsibility. Here's what successful companies are doing right now. They are using convergences and strategic foresight to define their strategy. All right, this is not either or. It's the whole job and sometimes it's a difficult thing to do, but you have to do it. You have to ask and answer where is the world

going? Where will value be created? And how will we participate all year long? This Is your job going forward as a leader. And for everybody else in this room, here's your personal plan. Use creative destruction on yourself. I want you to do some serious self-reflection. What are you doing out of habit that is already obsolete? What skill are you not building because it's hard? The augmented, emotionally resilient, irreplaceable person that you are in 2031 is being built right now by the choices that you're making. Storms are coming. You can't just stand outside watching the sky

turn green, watching the rain come in and go back indoors and say, "They're exaggerating. It's not going to be that bad. We'll be fine." Nobody's coming to save you. All right? If you want agency, you have to take action. And that starts today. [cheering] [applause] I made you a promise at the beginning that you would get new perspectives, new Processes, a new product, and new power. This is your new power folder. Um, I actually for the first time put my in most of my presentation, not the videos. You now have it. Um, and I because

because we have to get out there and do something if we're going to make positive change. All right. So this is it. Download it. It has everything that you need to get started. There's also hopefully live a new interactive tool on our website to see where you are in the World of change and convergence. If you want to chat, I will be at the bookstore right after this. Um, so you can come on by. And finally, finally, please look when you This has been a a great year for Southby, but also a tough year because

of logistics. So, I want you to find anybody with a Southby t-shirt on or a South staff or producer neck thing, and I mean everybody. And please, please, please tell them thankful. Uh, thank you. We are here. Yeah, [applause] we're all here thinking about the future, but they are literally taking care of us in the present. All right. Remember remember every civilization that has ever mattered was built by people like us. People who decided in the middle of uncertainty, in the middle of fear that we have agency. You have agency. I want you to go

out there and use it. Thank you. I will see you next year. [cheering] [applause]