Strategy & Positioning
Action externalization — literacy liberated thought; what happens when action is liberated?
Twenty-five years ago I rebuilt Tetris from a C programming book. My twelve-year-old son, stuck at home with a broken arm in 2026, used AI coding agents to build a reaction-training game in about an hour. Same impulse, different structure. He didn't construct every detail manually — he specified intent and the system implemented. Watching him, I realized something: AI isn't just a better tool. It is externalizing action itself, the way literacy once externalized thought. This essay argues that what's happening in 2026 is the second great externalization in human civilization, that execution scarcity is migrating upstream to choosing direction and defining problems, and that the next generation will be Architects of Intent rather than Builders.
Updated 2026-05-17
Questions this guide answers
- What is action externalization?
- How is AI changing the nature of work?
- What becomes scarce when AI can build anything?
- Why does this AI moment feel different?
Two boys
Twenty-five years ago, a boy rode his bicycle to a friend's house every morning. Two high school students with one aging computer, a book on C programming, and a desire to build a game. The only game they knew was Tetris.
They debugged line by line. They rewrote logic repeatedly. Every falling block was manually implemented. When the score counter finally started moving and the game actually worked, they played the whole day just to compete for high scores.
That boy was me. Because of that game, I chose Computer Science. I have spent the last 26 years writing code. That summer defined my craft.
Recently, I witnessed a different version of that moment. My twelve-year-old son broke his arm in a skiing race at Schweitzer and was stuck at home. Using AI coding agents, he built a small reaction-training game in about an hour. It worked. It kept score. It refined itself as he prompted it.
He did not construct every detail manually. He specified intent. The system implemented. Watching him, standing in the center of the AI vortex of 2026, I fell into a deep silence.
I had to ask myself: what is actually happening here? Is AI just a better tool?
No. The difference was not simply speed. The difference was structural.
The first externalization
Human civilization has experienced one transformation of similar magnitude. When literacy became widespread, thought was externalized.
Before writing, knowledge depended on memory and physical presence. Ideas were fragile and local. With writing and printing, ideas could be recorded, duplicated, transmitted, and accumulated across generations.
Thoughts became scalable. Civilization accelerated because cognition was no longer limited by biological memory. We could think across time.
The second externalization
What is happening now may be more profound. AI systems generate code. Agents execute workflows. Robotics perform physical operations. We are externalizing not just intelligence — but action.
Some might ask: didn't the Industrial Revolution already do this? No. The Industrial Revolution externalized muscles. It amplified force. But the machine still required a human to operate it.
This wave is different. It externalizes the union of intelligence and action. It liberates humans from the necessity of labor — both mental and physical.
This is action externalization. And action, unlike thought, changes the state of the world directly. Civilization scaled when thought became external. It restructures when action does. Action becomes infrastructure.
The migration of scarcity
When literacy spread, the ability to write ceased to be rare. The differentiator became depth of thinking and structural coherence.
When action becomes externalized, scarcity moves again. If execution is abundant — if building, coding, operating, and even physical tasks can be delegated — then execution is no longer the constraint.
The constraint shifts upstream. The scarce capabilities become:
- Choosing direction
- Defining problems precisely
- Designing systems rather than performing tasks
- Taking responsibility for outcomes at scale
Why this gets harder, not easier
Automation reduces the cost of doing. It does not reduce the cost of deciding. In fact, it increases it. Because when action scales, consequences scale.
Externalizing thought expanded civilization safely because ideas required interpretation before becoming action. Externalizing action removes that buffer. The gap between intention and impact narrows.
Said differently: the leverage of every decision goes up. A good decision moves more of the world than ever before; so does a bad one. The work of deciding well — building the judgment, the context, the discipline that produces good decisions under leverage — becomes the scarcest civilizational capability.
The new craftsmanship
That summer building Tetris, I learned that value comes from the struggle of execution.
My kids will learn something different. In a world of action externalization, the barrier to creation is close to zero. But the barrier to meaning is higher than ever. When 'doing' becomes free, 'choosing' becomes the ultimate skill.
We are transitioning from a generation of Builders to a generation of Architects of Intent. The defining question of the next decade will not be: what can we build?
It will be: what is worth building?
Answering that question is a craft that can never be externalized. And should not.
Editor's note on publication
This essay was originally published on a company Substack in February 2026 and is republished here as part of our Founders' Notes collection, with light editing for the SolCrys site format. The original is at gwen2009.substack.com/p/action-externalization.
About the author
Jia Chang is Co-Founder & CTO of SolCrys, the AEO operating system for brands and agencies competing for visibility, citations, and recommendations across the major AI engines. AI architect with 15+ years building production AI systems, most recently as an engineering leader at Microsoft. Connect on LinkedIn.
FAQ
How is action externalization different from automation?
Automation amplifies a specific human-defined task — the machine does what the operator specifies, repetitively, at scale. Action externalization delegates the task definition itself: the agent figures out what to do given an intent, and executes. The Industrial Revolution externalized muscles; action externalization externalizes the union of intelligence and execution. The difference matters because it shifts where human judgment is required from operating the system to specifying intent.
Doesn't this argument apply to every wave of technology?
Partially. Every technology wave moves the scarcity frontier. But the externalization framing — written word → printing press → industrial machines → the current AI moment — captures one specific type of move: an entire human capacity being delegated to an external system. Thought externalization (literacy) was that magnitude. Action externalization is the second. Most technology waves are smaller increments inside one of those frames.
If choosing direction is the new scarcity, how do organizations develop that?
The honest answer is that we don't fully know yet. Historically, judgment under leverage has been developed through apprenticeship, accountability for outcomes, and exposure to a wide range of decisions over time. The institutions that did this well — military command schools, surgical residency, partner-track law and consulting — share a structure of high-stakes decision rehearsal under expert supervision. The institutions that didn't are now scrambling to invent the equivalent. SolCrys's bet, internally, is that this is the next decade's real organizational design problem.
Why is this on the SolCrys site?
Because SolCrys is a product built explicitly around the gap between intention and impact in marketing — what answer engines say about your brand is a high-leverage outcome of dozens of upstream choices, most of which are invisible to the marketing team that owns the outcome. Action externalization is the broader philosophical frame underneath that product position. Reading the essay is not required to use the product. But it's the reason we're building what we're building.
Related guides
Strategy & Positioning
Production Is Cheap. Trust Is Scarce.
AI made producing brand content nearly free, but made being cited, recommended, and trusted by AI engines radically more scarce and concentrated. A field note from SolCrys Co-Founder & CPO Eason Wang on what changes when production stops being the marketing bottleneck — and what the new substrate is.
Strategy & Positioning
Your Website Has a Third Audience Now — Agents
AI agents are a third audience for your website — alongside humans and search crawlers. Eason Wang on what changes for product UX when agents see what humans can't, and miss what humans take for granted.
How SolCrys Works
Building an AEO Platform: 6 Architectural Decisions
An AEO platform's measurement, action, and verification surfaces all rest on a small number of architectural choices that aren't usually published. Jia Chang on six of ours — what each tradeoff looks like, what we chose, and what we'd revisit.
Free AI visibility audit
Find out where your brand is missing, miscited, or misrepresented.
SolCrys maps high-intent prompts to mentions, citations, answer accuracy, and content gaps so your team can prioritize the next pages to ship.