Simon 同时指挥三四个编码 agent，成了 30 到 40 倍的工程师；别人过不去，卡在上下文散落各处和工作无法验证。
个人：从自行车到汽车
最早的苗头出现在知识工作的大祭司身上：程序员。
我的联合创始人 Simon 曾是我们说的那种 10× 程序员，但他现在很少写代码了。走过他的工位，你会看到他同时编排三四个 AI 编码 agent，而它们不只是打字更快，它们会思考，合起来让他成了 30-40× 的工程师。他在午饭前或睡前把任务排进队列，让它们在他离开时干活。他成了无限心智的管理者。
1970 年代《科学美国人》一项关于移动效率的研究，启发了 Steve Jobs 那个著名的"心智的自行车"隐喻。只不过从那以后的几十年里，我们一直在信息高速公路上蹬自行车。1980 年代，Steve Jobs 把个人电脑叫作"心智的自行车"。十年后，我们铺出了叫互联网的"信息高速公路"。但今天大多数知识工作仍然靠人力驱动。就像我们一直在高速公路上蹬着自行车。
有了 AI agent，像 Simon 这样的人已经从骑自行车升级成了开汽车。
其他类型的知识工作者什么时候能拿到汽车？有两个问题必须先解决。
和编码 agent 比，为什么 AI 帮通用知识工作更难？因为知识工作更碎片化，也更难验证。第一是 context 碎片化。做编码时，工具和上下文倾向于集中在一处：IDE、repo、终端。但通用的知识工作散在几十个工具里。想象一个 AI agent 要起草一份产品简报：它得从 Slack 线程里拉，从一份战略文档里拉，从仪表盘里上季度的指标里拉，还得从只存在于某个人脑子里的机构记忆里拉。今天，人是那块胶水，靠复制粘贴和在浏览器标签页之间切换把这些缝起来。在这些 context 被合并之前，agent 会一直卡在窄用例里。
第二个缺的东西是可验证性。代码有个神奇的属性：你能用测试和报错来验证它。模型厂商用这一点来训练 AI 把代码写得更好（比如强化学习）。但你怎么验证一个项目是不是管得好，或者一份战略备忘录是不是写得好？我们还没找到为通用知识工作改进模型的办法。所以人仍然必须在环里，去监督、引导、示范什么才叫好。
Notion's CEO uses three images from industrial history to describe how AI reshapes knowledge work, and independently names the snag: general knowledge work can't be verified.
Indigo's conclusion
The strongest point is Notion's CEO naming, independently and from the product side, verifiability as the top bottleneck for general knowledge-work agents. What needs saying: the other bottleneck, fragmented context, is exactly what he sells, so real insight and product story are tangled together.
How to read this A vision essay from Notion's CEO with a heavy stake: both of its core bottlenecks are problems Notion sells a solution to. Read the genuine insider insight and the product story separately. Notion's internal numbers are credible; discount the grand vision and the timelines.
What to remember
He has no stake in the verification problem and a big one in the context problem; the two bottlenecks don't deserve equal credence.
700 agents beside 1,000 employees is real data, but they do meeting notes, IT and weekly reports: work that is cheap to verify.
Three images: individuals from bicycle to car; organizations with steel and steam still swapping water wheels; the economy from Florence to Tokyo.
Breakdown · 4 steps
01
Whoever masters the miracle material defines the era
Steel, semiconductors, now infinite minds. The hard part is that the future disguises itself as the past: making AI a Google search box is driving by the rear-view mirror. Read this part →
02
Trading the bicycle for a car takes solving two problems
Simon runs three or four coding agents at once and became a 30-to-40x engineer. Others can't follow, stuck on scattered context and work that can't be verified. Read this part →
03
AI is the steel of organizations, but we're still swapping water wheels
Human communication no longer has to be the load-bearing wall. Factory owners first just swapped water wheels for steam engines, with modest gains. Notion already runs 700+ agents beside its 1,000 employees. Read this part →
04
The knowledge economy goes from Florence to Tokyo
Steel and steam made cities explode at the cost of legibility. Organizations, he says, will be the same: trading some clarity for scale and speed. Read this part →
What would change my mind
Notion's agents start carrying load-bearing decisions that can't be verified, strategy and judgment, not just repetitive work.
How to read this
A vision essay from Notion's CEO with a heavy stake: both of its core bottlenecks are problems Notion sells a solution to. Read the genuine insider insight and the product story separately. Notion's internal numbers are credible; discount the grand vision and the timelines.
Whoever masters the miracle material defines the era
Steel, semiconductors, now infinite minds. The hard part is that the future disguises itself as the past: making AI a Google search box is driving by the rear-view mirror.
Every era is shaped by its miracle material. Steel forged the Gilded Age. Semiconductors switched on the Digital Age. Now AI has arrived as infinite minds. If history teaches us anything, those who master the material define the era.
There is a figure here in the original · See it in the originalLeft: teenage Andrew Carnegie and his younger brother.
There is a figure here in the original · See it in the originalRight: Pittsburg steel factories during the Glided Age.
In the 1850s, Andrew Carnegie ran through muddy Pittsburgh streets as a telegraph boy. Six in ten Americans were farmers. Within two generations, Carnegie and his peers forged the modern world. Horses gave way to railroads, candlelight to electricity, iron to steel.
Since then, work shifted from factories to offices. Today I run a software company in San Francisco, building tools for millions of knowledge workers. In this industry town, everyone is talking about AGI, but most of the two billion desk workers have yet to feel it. What will knowledge work look like soon? What happens when the org chart absorbs minds that never sleep?
Early movies often looked like stage plays, with one camera focused on the stage.This future is often difficult to predict because it always disguises itself as the past. Early phone calls were concise like telegrams. Early movies looked like filmed plays. (This is what Marshall McLuhan called "driving to the future via the rearview window.")
The most popular form of AI today look like Google search of the past. To quote Marshall McLuhan: "we are always driving into the future via the rearview window."Today, we see this as AI chatbots which mimic Google search boxes. We're now deep in that uncomfortable transition phase which happens with every new technology shift.
I don't have all the answers on what comes next. But I like to play with a few historical metaphors to think about how AI can work at different scales, from individuals to organizations to whole economies.
02
Trading the bicycle for a car takes solving two problems
Simon runs three or four coding agents at once and became a 30-to-40x engineer. Others can't follow, stuck on scattered context and work that can't be verified.
Individuals: from bicycles to cars
The first glimpses can be found with the high priests of knowledge work: programmers.
My co-founder Simon was what we call a 10× programmer, but he rarely writes code these days. Walk by his desk and you'll see him orchestrating three or four AI coding agents at once, and they don't just type faster, they think, which together makes him a 30-40× engineer. He queues tasks before lunch or bed, letting them work while he's away. He's become a manager of infinite minds.
A 1970s Scientific American study on locomotion efficiency inspired Steve Jobs's famous 'bicycle for the mind' metaphor. Except we've been pedaling on the Information Superhighway for decades since.In the 1980s, Steve Jobs called personal computers "bicycles for the mind." A decade later, we paved the "information superhighway" that is the internet. But today, most knowledge work is still human-powered. It's like we've been pedaling bicycles on the autobahn.
With AI agents, someone like Simon has graduated from riding a bicycle to driving a car.
When will other types of knowledge workers get cars? Two problems must be solved.
Comparing with coding agent, why is it more difficult for AI to help with knowledge work? Because knowledge work is more fragmented and less verifiable.First, context fragmentation. For coding, tools and context tend to live in one place: the IDE, the repo, the terminal. But general knowledge work is scattered across dozens of tools. Imagine an AI agent trying to draft a product brief: it needs to pull from Slack threads, a strategy doc, last quarter's metrics in a dashboard, and institutional memory that lives only in someone's head. Today, humans are the glue, stitching all that together with copy-paste and switching between browser tabs. Until that context is consolidated, agents will stay stuck in narrow use-cases.
The second missing ingredient is verifiability. Code has a magical property: you can verify it with tests and errors. Model makers use this to train AI to get better at coding (e.g. reinforcement learning). But how do you verify if a project is managed well, or if a strategy memo is any good? We haven't yet found ways to improve models for general knowledge work. So humans still need to be in the loop to supervise, guide, and show what good looks like.
The Red Flag Act of 1865 required a flag bearer to walk ahead of the vehicle while it drove down the street (repealed in 1896). An example of undesirable "human in the loop."Programming agents this year taught us that having a "human-in-the-loop" isn't always desirable. It's like having someone personally inspect every bolt on a factory line, or walk in front of a car to clear the road (see: the Red Flag Act of 1865). We want humans to supervise the loops from a leveraged point, not be in them. Once context is consolidated and work is verifiable, billions of workers will go from pedaling to driving, and then from driving to self-driving.
03
AI is the steel of organizations, but we're still swapping water wheels
Human communication no longer has to be the load-bearing wall. Factory owners first just swapped water wheels for steam engines, with modest gains. Notion already runs 700+ agents beside its 1,000 employees.
Organizations: steel and steam
Companies are a recent invention. They degrade as they scale and reach their limit.
Organizational chart for the New York and Erie Railroad, 1855. The modern corporation and org chart evolved with the railroad companies, which were the first enterprises that needed to coordinate thousands of people across great distances.A few hundred years ago, most companies were workshops of a dozen people. Now we have multinationals with hundreds of thousands. The communication infrastructure (human brains connected by meetings and messages) buckles under exponential load. We try to solve this with hierarchy, process, and documentation. But we've been solving an industrial-scale problem with human-scale tools, like building a skyscraper with wood.
Two historical metaphors show how future organizations can look differently with new miracle materials.
A wonder of steel: the Woolworth building was the tallest building in the world upon completion in NYC, 1913.The first is steel. Before steel, buildings in the 19th century had a limit of six or seven floors. Iron was strong but brittle and heavy; add more floors, and the structure collapsed under its own weight. Steel changed everything. It's strong yet malleable. Frames could be lighter, walls thinner, and suddenly buildings could rise dozens of stories. New kinds of buildings became possible.
AI is steel for organizations. It has the potential to maintain context across workflows and surface decisions when needed without the noise. Human communication no longer has to be the load-bearing wall. The weekly two-hour alignment meeting becomes a five-minute async review. The executive decision that required three levels of approval might soon happen in minutes. Companies can scale, truly scale, without the degradation we've accepted as inevitable.
A mill with a water wheel to power its operations. Water was powerful but unreliable and restricted mills to a few locations and seasonality.The second story is about the steam engine. At the beginning of the Industrial Revolution, early textile factories sat next to rivers and streams and were powered by waterwheels. When the steam engine arrived, factory owners initially swapped waterwheels for steam engines and kept everything else the same. Productivity gains were modest.
The real breakthrough came when factory owners realized they could decouple from water entirely. They built larger mills closer to workers, ports, and raw materials. And they redesigned their factories around steam engines (Later, when electricity came online, owners further decentralized away from a central power shaft and placed smaller engines around the factory for different machines.) Productivity exploded, and the Second Industrial Revolution really took off.
This 1835 engraving by Thomas Allom depicts a textile factory in Lancashire, UK. It was powered by steam engines.We're still in the "swap out the waterwheel" phase. AI chatbots bolted onto existing tools. We haven't reimagined what organizations look like when the old constraints dissolve and your company can run on infinite minds that work while you sleep.
At my company Notion, we have been experimenting. Alongside our 1,000 employees, more than 700 agents now handle repetitive work. They take meeting notes and answer questions to synthesize tribal knowledge. They field IT requests and log customer feedback. They help new hires onboard with employee benefits. They write weekly status reports so people don't have to copy-paste. And this is just baby steps. The real gains are limited only by our imagination and inertia.
04
The knowledge economy goes from Florence to Tokyo
Steel and steam made cities explode at the cost of legibility. Organizations, he says, will be the same: trading some clarity for scale and speed.
Economies: from Florence to megacities
Steel and steam didn't just change buildings and factories. They changed cities.
Until a few hundred years ago, cities were human-scaled. You could walk across Florence in forty minutes. The rhythm of life was set by how far a person could walk, how loud a voice could carry.
Then steel frames made skyscrapers possible. Steam engines powered railways that connected city centers to hinterlands. Elevators, subways, highways followed. Cities exploded in scale and density. Tokyo. Chongqing. Dallas.
These aren't just bigger versions of Florence. They're different ways of living. Megacities are disorienting, anonymous, harder to navigate. That illegibility is the price of scale. But they also offer more opportunity, more freedom. More people doing more things in more combinations than a human-scaled Renaissance city could support.
I think the knowledge economy is about to undergo the same transformation.
Today, knowledge work represents nearly half of America's GDP. Most of it still operates at human scale: teams of dozens, workflows paced by meetings and email, organizations that buckle past a few hundred people. We've built Florences with stone and wood.
When AI agents come online at scale, we'll be building Tokyos. Organizations that span thousands of agents and humans. Workflows that run continuously, across time zones, without waiting for someone to wake up. Decisions synthesized with just the right amount of human in the loop.
It will feel different. Faster, more leveraged, but also more disorienting at first. The rhythms of the weekly meeting, the quarterly planning cycle, and the annual review may stop making sense. New rhythms emerge. We lose some legibility. We gain scale and speed.
Beyond the waterwheels
Every miracle material required people to stop seeing the world via the rearview mirror and start imagining the new one. Carnegie looked at steel and saw city skylines. Lancashire mill owners looked at steam engines and saw factory floors free from rivers.
We are still in the waterwheel phase of AI, bolting chatbots onto workflows designed for humans. We need to stop asking AI to be merely our copilots. We need to imagine what knowledge work could look like when human organizations are reinforced with steel, when busywork is delegated to minds that never sleep.
Steel. Steam. Infinite minds. The next skyline is there, waiting for us to build it.
3:55 AM · Dec 23, 2025·6.2M
Where Indigo landsFurther
Indigo's conclusion
The strongest point is Notion's CEO naming, independently and from the product side, verifiability as the top bottleneck for general knowledge-work agents. What needs saying: the other bottleneck, fragmented context, is exactly what he sells, so real insight and product story are tangled together.
What to remember
He has no stake in the verification problem and a big one in the context problem; the two bottlenecks don't deserve equal credence.
700 agents beside 1,000 employees is real data, but they do meeting notes, IT and weekly reports: work that is cheap to verify.
Three images: individuals from bicycle to car; organizations with steel and steam still swapping water wheels; the economy from Florence to Tokyo.
Claims you can check later
Claim
Who
When we will know
How firm
Once context is consolidated and work is verifiable, billions of workers go from pedaling to driving to self-driving
Ivan Zhao
Trend
First-hand judgment; heavy self-interest
AI is the steel of organizations; a two-hour weekly alignment meeting becomes a five-minute asynchronous review
Ivan Zhao
Near future
First-hand vision; product story
The knowledge economy goes from Florence to Tokyo: thousands of agents and people running continuously across time zones
Ivan Zhao
Medium term
First-hand vision
The rhythm of weekly meetings, quarterly planning and annual reviews may stop holding
Ivan Zhao
Medium term
First-hand judgment
Back on the long-running theses
confirms
Verification can't be compressed: bottleneck, moat, breaking point Ivan is a fourth independent source, after Tara, Google's data and Noam, pointing to the same scope.
conflicts + adds to
An enterprise Markdown OS: nobody is building it yet He describes exactly this view's end state, but the problem is defined and the solution supplied by an interested party; delivery is in doubt.
adds to
The AI jobs apocalypse: replacement task by task Frontline testimony from inside Notion: replacement starts with checkable, low-risk repetitive work.
adds to
You don't own the model People move up from inside the loop to a supervising position with leverage; value shifts to taste, judgment and showing what good looks like.