Predictions

This is where I maintain a list of my predictions about technology and society—along with their accuracy over time.

I do this because I'm obsessed with improving my model for how the world works, and there's no better test than making clear predictions and watching them fall apart over time (or not).

The goal is to find systemic errors in my thinking that I can then correct.

Main/Topical Predictions ​

AI-Powered Digital Assistants ​

Digital Assistants as Primary Interface: In 2016 I wrote The Real Internet of Things where I predicted that AI-powered digital assistants would become our primary interface with technology - not just voice assistants but tireless advocates working 24/7 on our behalf. We're now seeing this with ChatGPT and Claude becoming exactly these kinds of assistants.

The API Economy ​

Everything Becomes an API: I also said in that book that businesses and objects would become APIs - essentially daemons that our digital assistants would interact with directly. This is now happening with technologies like MCP (Model Context Protocol) where AI agents can directly connect to business services through Universal Daemonization protocols.

Augmented Reality as Display Layer ​

AR Glasses for Contextual Overlays: Another prediction from that same book was that Augmented Reality would become the display layer for all this data - AR glasses providing contextual overlays on the real world based on Continuous Customization of our preferences. Meta and Apple are now racing to build exactly these kinds of glasses.

The Meaning Crisis ​

Finding Purpose in Virtual Worlds: Back in 2006, I predicted that people would increasingly lack meaning and look for it in games and virtual worlds as AI handles more cognitive work. The explosion of gaming culture and VR experiences suggests this trend is well underway, as I explore in The EM Wave of Crisis and Appreciation.

Ambient Personalization ​

Preference Broadcasting and Adaptive Environments: In 2014, I predicted that personal daemons would broadcast our preferences and that venues would personalize themselves based on customer preferences through Continuous Customization. Smart home devices and retail personalization are early examples of this trend, as detailed in my Daemon Personal API project.

Material Progress vs Human Fulfillment ​

The Pinker Paradox: In 2018, I critiqued Steven Pinker's Enlightenment Now, arguing that despite material progress, people would feel increasingly unhappy due to lack of meaning and purpose as AI takes over more human work (see The Future of Work section in my IoT article). Seven years later in 2025, this seems to be exactly what's happening - we have unprecedented prosperity yet rising anxiety and depression rates.

Current Predictions Table ​

Here's a short view of the predictions in table format.

Correct Predictions ​

Missed Predictions ​

Predictions by Category ​

Technology & AI ​

AGI Timeline ​

• Prediction: We'll have AGI between 2025 and 2028
• Confidence: Almost Certain
• Date Predicted: March 2023
• Status: 🔄 In Progress
• Notes: I believe the combination of scaling, new architectures, and "tricks" will get us there faster than most expect.

Open Source AI ​

• Prediction: Open-source models will largely catch up to closed-source models
• Confidence: Probable
• Date Predicted: February 2023
• Status: 🔄 In Progress
• Notes: The gap is already narrowing significantly with models like Llama and others.

Apple's AI Implementation ​

• Prediction: Apple is about to go from the worst AI implementation to the best
• Confidence: Probable
• Date Predicted: January 2025
• Status: 🔄 In Progress
• Notes: Apple's focus on privacy-first, on-device AI will differentiate them significantly.

Universal APIs ​

• Prediction: Everyone, including people, will have an API
• Confidence: Almost Certain
• Date Predicted: December 2016
• Status: 🔄 In Progress
• Notes: MCP and similar protocols demonstrate this trend.

Society & Economics ​

AI Job Displacement ​

• Prediction: Recession-like shock caused by AI job loss by 2027
• Confidence: Chances About Even
• Date Predicted: July 2025
• Status: 🔄 In Progress
• Notes: The pace of AI development suggests significant economic disruption is inevitable.

Meaning Crisis ​

• Prediction: People will lack meaning and look for it in games
• Confidence: Almost Certain
• Date Predicted: July 2006
• Status: 🔄 In Progress
• Notes: This trend is already visible with the rise of gaming, VR, and immersive experiences.

Personal Technology ​

Preference Broadcasting ​

• Prediction: Personal daemons will broadcast preferences
• Confidence: Almost Certain
• Date Predicted: March 2014
• Status: 🔄 In Progress
• Notes: Smart devices and AI assistants are beginning to share context between services.

Venue Personalization ​

• Prediction: Venues personalize based on customer preferences
• Confidence: Almost Certain
• Date Predicted: March 2014
• Status: 🔄 In Progress
• Notes: Early examples exist in hospitality and retail, will accelerate with AI.

Geopolitics ​

Russia Trading Status ​

• Prediction: Russia will significantly return to normal trading status by 2027
• Confidence: Chances About Even
• Date Predicted: March 2023
• Status: 🔄 In Progress
• Notes: This depends heavily on the resolution of the Ukraine conflict.

Analysis ​

Observation 1: Trump and Russia ​

One thing I seem to have been most wrong about is the behavior of Trump in relation to Russia. I think this is a good learning opportunity for me.

My assumption was that Trump's greed would be a giant target and he would be easily manipulated into capitulating completely against Ukraine and completely for Russia. And what has happened instead is a kind of a mush of a little bit of Russian opposition and a little bit of support of Ukraine.

I will continue thinking about what I can learn from this. Perhaps one thing is simply that Trump is not very predictable. But that doesn't feel like a lesson I can do much with, Unless I could figure out how to determine this again in the future for him or others.