Agents

In short: AI agent cost optimization starts with context growth. A long-running agent can move from $50 a month to $2,500 because each turn resends system prompts, tool definitions, memory, files, and earlier messages. Four practices bring the bill back under control: prompt caching, lazy-loaded tools, model routing, and context cleanup. When an agent is new, the system prompt may be only a few hundred tokens, with two or three tools. Then the prompt grows, the tool list expands, memory accumulates, and every turn starts paying for earlier turns. The Claude system prompt leaked in late 2024 was 24,000 tokens, nearly 50 times larger than the starting point. OpenClaw users have reported sending more than 150,000 input tokens to Gemini 3.1 Pro, only to get 29 output tokens in the first turn. An unoptimized agent handling 100 messages a day with 166K input tokens can cost about $996 a month on Gemini 3.1 Pro and about $2,490 a month on Claude Opus 4.6. There are ways to push that cost back down to $50-$100 a month. ...

In short: AI agents do more than automate work. As humans domesticate AI with prompts, evals, and workflows, AI is also domesticating us by taking over the first move: the first outline, the first judgment, the first messy sentence, the first uncomfortable question. Wheat, Humans, and the Direction of Domestication In Sapiens: A Brief History of Humankind, Yuval Noah Harari makes a slightly uncomfortable point about the agricultural revolution: perhaps humans did not domesticate wheat so much as wheat domesticated humans. It sounds like a clever reversal at first, but the accounting is fairly plain. Wheat started as a wild grass in the Middle East. Over time it spread across the world, occupied enormous amounts of land, and got humans to clear fields, bend their backs, pull weeds, dig channels, build granaries, and stop wandering. Wheat did well. Human backs, less so. That story is useful before talking about AI, because it cuts through a lot of vague language about technology changing the world. A tool is not always something you use and then put back on the table. Stay with it long enough and it starts changing your movements, your schedule, and your sense of what feels normal. Wheat changed posture and settlement. The internet changed attention. AI is reaching a little further inward. It is changing how we begin to think about things. ...

In short: AI coding agents are changing the status of code. In mature agentic workflows, code is no longer only written by humans; it is repeatedly generated, tested, corrected, and selected by an optimization loop. That makes tests look like loss functions, production failures look like generalization failures, architecture look like inductive bias, and harness engineering look like optimizer design. Introduction: A Shift We Have Not Yet Named Precisely Over the past eighteen months, software development has undergone a quiet but forceful restructuring. Tools such as Cursor, Claude Code, and Codex are pushing us away from the old workflow of “humans write code, machines assist with completion” toward something structurally different: humans describe intent, define constraints, and provide feedback, while agents repeatedly generate, run, and revise code until some convergence condition is met. Most industry commentary still frames this shift in productivity terms: “AI makes us write code N times faster.” That framing misses a more basic ontological question: in this new workflow, what has happened to the nature of code itself? ...

In short: Many AI agents look productive but are actually drifting — confidently executing the wrong moves on a wrong picture of the situation. The bottleneck for the next phase of agent systems is not larger context windows or stronger base models; it is whether the system can construct and maintain a stable belief state. This piece argues why belief state quality is the right optimization target, proposes five proxy metrics to measure it, and lays out where to put incremental engineering resources next. AI agents that look productive often turn out to be drifting — confidently executing the wrong moves on a wrong picture of the situation. Competition in agent systems is shifting from “whose model is stronger” toward “who can keep producing higher-quality belief state.” If you accept that framing, several seemingly unrelated problems suddenly line up: the same model behaves very differently inside different product shells; long-running agents fail not because they cannot answer but because their judgment of the situation is wrong; context windows keep growing, but system capability does not scale linearly with them; and scattered engineering pieces — skill, memory, retrieval, tool use, trace, summary — all start to matter at the same time. ...

A friend recently showed me his new tool, beaming with excitement. He follows about 5,000 people on X. Researchers, founders, investors, developers, media figures — after years of accumulating, his feed had long since become a bottomless waterfall. He’d tried “read later” apps before, bookmarking over a thousand articles and actually reading five. Like most people. Now he uses an AI agent that reads the full output of all 5,000 accounts, compressing everything into 20 curated highlights per day. Fifty-four structured briefings in ten days. What used to take two hours to skim now takes five minutes. Ninety-five percent of noise, filtered out. “The root of information anxiety is the cost of filtering,” he said. “Hand the filtering to an agent, and the anxiety disappears.” He’s right. But only about the first half. The anxiety does disappear. What also disappears is everything you didn’t know you needed to know. Five thousand tweets compressed to twenty. Among the 4,980 discarded, there might have been one from a field you’ve never followed, using logic you’ve never encountered, explaining a problem you thought you’d already figured out. ...

Most people came to GTC 2026 expecting new GPUs. What they got instead was something much bigger: AI is no longer being presented as software. It is being redefined as infrastructure. This shift shows up everywhere: from NVIDIA’s “AI factory” framing to the rise of agent-based systems like OpenClaw. If you still think of AI as a tool or a feature, you are looking at the wrong layer. What’s being built now is a new kind of computing system, not software. AI Factories Are Not a Metaphor NVIDIA’s idea of “AI factories” is easy to misunderstand. It sounds like a bigger data center, but that framing misses the point. A traditional data center stores and processes data. An AI factory produces something else entirely: intelligence, in the form of tokens, decisions, and actions. In other words: Input: data Output: tokens System: large-scale coordinated compute AI factories produce intelligence the way factories produce goods. This is a structural shift. Data centers used to be part of IT. AI factories start to look more like industrial systems. ...

Technological revolutions rarely announce themselves. The agricultural revolution had no press release. The industrial revolution had no countdown. Even the internet only became obvious in hindsight. Artificial General Intelligence will likely arrive the same way. There will be no moment when the world collectively agrees that AGI has appeared. No headline. No global notification. Instead, there will only be a moment years later when people look back and say: That was when everything started to change. By then, the transformation will already be underway. And this time, we may have far less time to adapt. The Speed of This Revolution Technological revolutions have always accelerated. When the steam engine entered factories in the late 18th century, it began replacing manual labor. Yet Britain did not pass its first meaningful labor protection law until 1833—almost seventy years later. The Second Industrial Revolution moved faster. Electricity, steel, and chemical industries reshaped entire economies within decades. Germany transformed from an agrarian country into an industrial power in less than thirty years. The internet accelerated things again. ...

Something strange is happening in software. We can now ask an AI agent to implement a feature in minutes. Ship multiple builds in a single day. But submitting that build to Apple for signing still takes an hour. Code has taken off like a rocket. Everything around it is still crawling on the ground. The reason is simple: Coding has crossed a singularity. Recently a tool called OpenClaw went viral among developers for enabling agent-driven coding workflows. But if all you see is OpenClaw, you’re missing the real story. OpenClaw is not the story. It is a signal. A signal that something fundamental has changed in how software is created. Once you see that change clearly, a much deeper question appears: What happens to the world when coding stops being scarce? 1. The Most Important Change of 2026 For decades, the software industry operated under one basic assumption: Coding ability is scarce. Code had to be written line by line. Systems had to be built gradually by teams. Engineering time was the most expensive resource in the company. ...

All of human civilization has always followed the same underlying structure: ideas are abundant, but execution is what creates value. For most of history, the ability to get things done determined who won and who didn’t. Everyone knows what kinds of activities are considered useful—working out, learning a foreign language, reading, building products, starting projects. And everyone also knows this: wanting to do something is rarely the bottleneck. The real constraint has always been execution. Companies are built around execution. Management exists to keep execution from going off track. Salaries exist to make people willing to execute. Education exists to give people the ability to execute. Venture capital invests in execution as well. You have an idea, I have an idea—who gets the money? The one who can make it real. After Agents, a single person with a single weekend can build what previously required an entire team working for half a year. Everyone now has nearly unlimited execution power. We have entered the age of spectacle. At this moment, “getting things done” has shifted from being a scarce resource to basic infrastructure. And once that happens, we are forced to rethink the question of value: what, exactly, is still worth something? ...

Introduction Most of what we learned in the Internet era is no longer compounding. DAU is losing relevance. SaaS is no longer the growth engine it once was. The attention economy is in structural decline. The classic path from tools to platforms is breaking down. The term “AI application” no longer describes what is actually being built. Network effects. Communities. Platforms. SaaS. Applications. Attention economy. These concepts once formed a shared framework for understanding technology and business. We used them to design products, explain strategy, and communicate with investors. But more and more often, it becomes clear that the world these concepts describe is no longer the center of gravity. Not because the Internet suddenly disappeared, but because its core assumptions are no longer where growth comes from. The Internet era was built on one fundamental premise: Humans are the users of software. That premise is now eroding. A new one is taking its place: Agents are becoming the primary operators of software. This is not a sudden collapse. It is a gradual handover. ...