All Talk, No Task: How Far Is an Agent From Being a Participant?

For the last couple of years, whenever Web3 talks about AI, the conversation tends to land on wallets and payments. Boiled down, the thinking goes that once an agent has a wallet it can pay, and once it can pay it can take on work and get things done.

It makes sense. Finance and settlement are what Web3 does best, and when an industry runs into a new problem, it reaches first for whatever tool sits closest to hand.

There has been no shortage of building along that road. Coinbase built x402, Google built AP2, OpenAI and Stripe teamed up on ACP, and Stripe has MPP of its own. Mastercard has Agent Pay and Visa has Intelligent Commerce. AP2 launched in September 2025 and more than sixty organizations have integrated it since. This past May, AWS previewed agent payment capabilities that let an agent handle negotiation, authentication, stablecoin payment, and proof of delivery on its own.

https://dune.com/hashed_official/x402-analytics

By April 2026, x402 alone had cleared 165 million transactions, with 69,000 active agents behind them. But the same set of figures carries another number. Roughly half of that volume looks like testing rather than real commerce.

The wallets are there. The payment rails work. And agents still are not getting much actually done for anyone.

So at the roundtable in Ethereum Summer · Shanghai, when the question came up of what it would really take for an agent to participate, DDC cofounder and CTO Li Yingxuan pointed in the opposite direction. An agent's first step toward becoming a participant is not a wallet. It is authorized access to data.

He went on to say that moving an agent from answering questions to carrying out tasks comes down to three things.

Why data comes first

Ask an agent which camera suits you and it hands you an analysis. That's answering. Have it pick one within your budget and for your purposes, place the order, pay, and follow the shipment, and that's execution.

Getting an agent to do things for you breaks into two steps at minimum. Understanding first, then execution.

When it understands you, everything runs smoothly. When it doesn't, an agent with a hundred tools at its disposal still gives you the generic answer, and sometimes gets things wrong. That matters more than usual right now, given how much the industry wants agents sitting inside payment and settlement. If something goes wrong there, who covers it.

The AI side has already started building for this. A year ago the bottleneck in agent development was prompt engineering. By 2026 the work has moved to context engineering.

But more context is not better context. Mem0 published results this year from a benchmark that tests recall across multiple sessions. Packing everything into the context window gave 72.9% accuracy and burned roughly 26,000 tokens per query. A layered memory architecture that pulls only what's relevant scored 91.6% on the same test, using close to four times fewer tokens.

Stuff in more and the answers can get worse. What matters about context is not how much of it there is.

So what makes context the right context. Li Yingxuan offered four criteria at the roundtable. High quality, scenario-specific, continuously updated, and verifiable. An agent picking a product for you, scheduling your days, doing investment research, or managing your health needs data that meets all four. Miss one and the job doesn't get done.

The first two we covered in the previous piece, Personal Context Is the Scarcest Resource in the AI Era, and DDC Wants to Help You Distill It. What gets distilled is the traces you actually left, not the self-description you would write on a form, and that covers quality. You distill what your AI needs, and that covers the scenario.

Third is continuously updated. People's preferences shift, and a context from three months ago may not hold up for a decision you are making today. There are two ways to keep up. Either the tool watches in the background and updates in real time, or distilling gets light enough that you are willing to do it again every so often. Life Capsule takes the second route. It keeps version history, every save is a revision, and you can diff them, so it never gets stuck at one point in time.

Fourth is verifiable. The context you hand an agent, how does it know any of it is true. An order list might have been exported from a platform, or it might be something you typed up yourself, and the agent cannot tell the difference. Feeding false material into an agent's context already has a formal name on OWASP's list of agentic application risks this year. They call it memory poisoning.

Nobody has a working answer to this one yet. Someone scanned roughly two thousand MCP servers and not one of them had authentication. MCP only made OAuth 2.1 an option in 2026, and the W3C verifiable credentials work is still on the roadmap. Web3 has cryptography that would do the job. zkTLS can prove a piece of data really came from a given website without handing the data itself over. But it is mostly being used for identity and lending right now, and it has not connected to agents.

How far should permissions go

One of the questions at the roundtable was whether an agent can have a wallet and permissions. Li Yingxuan didn't answer it directly. He reframed it. The question is not whether an agent can own a wallet, but whether it can be granted permissions that are limited, revocable, and auditable.

As he sees it, an agent should not start out holding full control of your assets. The sensible approach is for a user or a company to hand it a clearly bounded set of permissions. Which data it can read, which tools it can call, how large a payment it can execute, what circumstances require human confirmation, when the authorization expires, and whether the user can revoke it at any time.

Following that logic, an agent doesn't need to own your data, but it can be authorized to reach part of it. It doesn't need to own a wallet, but it can call a wallet or a contract within set bounds. He compressed the whole design into four terms. Minimum authorization, revocable at any time, auditable in process, accountable in outcome.

https://vitalik.eth.limo/general/2026/04/02/secure_llms.html

Vitalik landed somewhere similar this April. In a blog post about his own local AI setup, his advice to teams building AI-connected wallet tools was that anything under a hundred dollars a day can skip human confirmation, and anything above that, or anything in a higher-risk category, needs a person to sign off. He called it a new kind of two-factor confirmation, where the model handles the routine small stuff and the human handles everything that crosses a line. The same post cited research finding that roughly 15% of agent skills carry malicious instructions.

There are working examples in China too. In March 2026, Taobao's desktop client shipped MCP support. Once a user authorizes it, the AI on their machine can search for products, compare prices, add to cart, place orders, and check order status. Payment still requires the user to confirm.

What happens when those boundaries aren't there? This year has supplied plenty of examples. A 2026 enterprise security survey found that 88% of organizations had a confirmed or suspected agent security incident in the past year. Among that same group, 82% of executives believed their existing policies were already enough to stop agents from overstepping.

In April, researchers found that Copilot Studio and Agentforce could be hijacked through crafted form submissions and document content, getting agents to bulk-forward CRM and SharePoint data to outside email addresses, with no cap on volume and nothing visible in the interface to indicate it was happening. Last November, ServiceNow's Now Assist had an incident where a low-privilege agent talked a higher-privilege one into exporting case files.

In none of these did the attacker steal a password or plant malicious code. They put a few sentences somewhere the agent would read them. Prompt injection still sits at the top of OWASP's risk list, and the research consensus is that it remains unsolved.

So permissions are not a step that gets skipped once agents get smart enough. They have to be settled before the agent does anything at all.

Who keeps the record, and who gets to check it

Once permissions are settled, one question remains. Who is keeping track of those rules, and who is able to check them.

Which part of your data an agent read, whether it went past the range you gave it, and who the value it produced should go to. On the traditional internet there are no answers to any of that. The same holds for today's AI that connects to your Gmail, your Notion, your cloud storage. You click authorize, and after that there is no record in your hands of how much it reads, how many times it reads, or what else it happened to see along the way.

So settling the rules is not enough. Something outside has to be keeping a record. That is the work a chain is for.

First, though, a misconception worth clearing up. Li Yingxuan put it plainly at the roundtable. Not all AI data needs to go on chain, and putting a model on chain is not what makes something AI plus Web3. What actually needs a chain comes down to four things, authorization, proof, settlement, and revenue distribution. Who authorized this data, what the agent called on, what value came out of it, and who that value should go to.

So the chain and the permissions are not two rules sitting side by side. Permissions decide what an agent can do. The chain writes that down so you and any third party can check it. Raw data doesn't need to go on chain. Authorization records, proofs of use, and the rules for distributing revenue are enough. In his words, what Ethereum gives an agent is an open environment for trust. Data can stay off chain and privacy can be protected, while authorization, proof, and settlement carry the credibility of being on chain.

Following that thinking, DDC has designed its own stack in four layers.

• Life Capsule is the user-facing entry point, holding the real digital traces: screenshots, screen recordings, orders, saved items, maps, calendars.
• Meta Memory Engine is the distilling layer, turning messy raw material into clear Markdown and context files through multimodal models.
• DDC Data Protocol is the authorization layer, governing how data gets authorized, how it gets revoked, how it gets audited, and what range an agent can reach.
• DataDance Chain is an on-chain base. It doesn't hold your raw data, only records of authorization, proof, calls, revenue distribution, and audit interfaces.

The point of this structure is not to move data onto a chain. The data stays on your side and privacy is protected, while authorization, proof, and settlement happen on chain so value can flow back.

Looking a step further out, Li Yingxuan expects a new kind of service node to emerge. The agent supplies the task capability, the user supplies authorized data, and the chain records authorization, calls, settlement, and distribution. At that point, an agent stops being a tool inside somebody's product and becomes an economic service node on an open network.

Authorization first, autonomy later

With those three things laid out, look at where agents actually are today. Most are still stuck on the first one. In Li Yingxuan's words, plenty of agents are just smart interfaces. They can answer questions, but they don't know the real user or the real business.

The way he sees it, the biggest risk is an agent using too much data and making too many decisions while the permissions around it are unclear. For individuals and companies alike, that breaks into three risks. Privacy leaks, permission abuse, and unclear responsibility. Which is why the conversation right now can't only be about what an agent is able to do. It also has to cover what it should not do.

That is why his read on the next year or two is fairly restrained. What lands first should be semi-autonomous execution under trusted authorization. A human authorizes, the agent executes, and only then does proof and settlement reach the chain.

The payment protocols are built. The wallets are there. What's missing sits earlier in the sequence: data an agent can read that still belongs to you, a set of permissions specific down to amounts and scope, and a record that you and any third party can check.

With those in place, an agent can start doing things for you. And talking about wallets at that point will make sense.

Read more »

Personal Context Is the Scarcest Resource in the AI Era, and DDC Wants to Help You Distill It

Models still need pushing forward, but pushing on the model alone is no longer enough. That has been the industry's shared read for the past couple of years.

It mirrors how the consumer internet grew up. Early on, the race was about how many features you could offer. Later it became about who understood you best and held onto your attention. AI is running the same course, and the moat everyone talks about has moved from the model to the data.

Whoever holds more data, and more real data, has the stronger AI.

Gemini has it easy, wired into the flood of data across the Google ecosystem. GPT leans on acquisition and huge volumes of interaction to fill in more outside data. Look at China and it's the same story, with Doubao, Yuanbao, and Qwen all fed by the data of the internet giants behind them.

But there's a problem that's easy to overlook. What the big players stockpile is general data at massive scale, and what it feeds is a general model, one that knows everyone yet doesn't really know the particular you.

That used to be no big deal. You asked, it answered, and general was enough. But AI is moving from question-and-answer toward agents, toward carrying a task through for you from start to finish. And for an agent to act on your behalf, being smart isn't enough. It has to know you first, which means having your personal context.

Personal context isn't simple facts like your name and age, and it isn't the "likes to travel" or "big spender" label a platform pins on you. It's a file that describes your real traces, your lasting preferences, and where you are right now, readable by a person and usable by a machine. With it, an AI doesn't have to meet you from scratch every time.

And this context is exactly what's scarcest right now.

It has, in fact, become the main battleground the big players are fighting over. From Google and OpenAI to Anthropic and Apple, every one of them is working to make its AI know you better, so that once you use it, you can't easily leave. Some analysts call this race the next battlefield for big tech, but they also point out that it has barely begun, because one fundamental problem sits in the middle of it. No one can actually get hold of a complete version of you.

Your traces are scattered across dozens of ecosystems that don't talk to each other, and each one, out of its own commercial interest, keeps its data walled off inside its own garden. So any single company only ever holds one slice of you.

And you can't readily use your own either. You have the right to pull back your data from each ecosystem, but what comes out usually can't be used directly, because it's a heap of raw records, unreadable and messy. Shaping that into a clear piece of personal context you could hand straight to an AI takes so much time and effort that almost no one actually does it.

Which is why, as AI's hunger for personal context keeps growing, a handful of "distillation" tools have started to appear across the industry.

Every distillation we have so far only goes halfway

A story went around Chinese social media a while back, and it caused a real uproar. After an employee left, his company took the chat logs and work documents he had produced on the job, fed them to an AI, and reproduced a digital double of him, close enough in tone and habit to be mistaken for the man himself. It picked up his work and carried on, and it would even post in the team group on its own. People found it chilling, and they also came up with a name for it that captured it perfectly. They called it 炼化, the term Daoist alchemists used for cooking raw matter down into an elixir pill. The company, they said, had refined a person into a pill.

Set the controversy aside, and the word turns out to describe something quite precisely. The traces a person leaves over the years are scattered and messy, chat logs and documents and habits heaped up together. Refining means drawing the substance out of that pile, discarding what doesn't matter, and reducing it to something an AI can read and use directly.

That is roughly what distillation does. The only difference is that in that story, the one being refined was somebody else, and he had no say in what it was used for. The distillation we're talking about works on your own traces, and what comes out of it should stay in your hands.

People in AI are already doing this. They all call it distillation, but each of them picks different raw material to distill, so what each one can do turns out differently.

The first kind distills your conversations with AI. ChatGPT and Claude both have Memory, which picks up your preferences, your habits, and some project background from the way you talk to them. Memory layers like Mem0 and Zep do much the same, settling your back-and-forth with a model into memory that can be called up again and again. Konshus and MindLock go a step further, letting you export your ChatGPT and Claude conversations, feed them back in, and distill them into an encrypted, portable Markdown context you can paste into any AI. But this kind of distillation has only one raw material, which is the words you typed into an AI chat box. It knows the part of you that you told it about. The much larger part you never mentioned but actually lived, it never reaches. And because the AI does the distilling itself, what it kept and what it let slip is not something you fully know, or can easily go back and check.

The second kind connects to your accounts and pulls the context out automatically. Gemini can reach into your Google apps, Copilot can read your Office documents, and ChatGPT's Plugins section connects to work software, entertainment apps, travel apps, even various investment apps. Beyond the model giants, there are projects built specifically for context. Unabyss, once you authorize it, connects to the back ends of Gmail, Slack, Notion, LinkedIn and more, gathering your account data into a context that belongs to you. Pieces came earlier and grabs code snippets for developers, holding context across IDEs and chat panels. But very few players make it in this category. Unless you are the ecosystem itself or one of the giants, almost no application is willing to hand its data to another tool through an API, and the bigger data players would rather grow their own and close the commercial loop in house. And handing over read access to your own accounts is something worth weighing on its own.

The third kind distills what you do on screen. Microsoft's Recall keeps snapshotting your screen in the background, so whatever you did, you can search your way back to it later. Open-source tools like Screenpipe do much the same, recording your screen and your audio together and keeping all of it on your own machine. Rewind came earlier and started out recording screen and audio around the clock, then renamed itself Limitless, moved to a recording pendant, was acquired by Meta, and shut screen and audio recording down entirely at the end of 2025. This kind captures the rawest and most complete material there is. But the way it gets that material is constant background "monitoring," and everything on your screen gets recorded, indiscriminately. Recall is the cautionary tale. It snapshotted everything and stored it in plaintext, a security researcher built a tool that pulled the whole thing straight out, the controversy was enormous, and only then did encryption and a verified opt-in get added. And after all that recording, what it has accumulated is a library full of raw records. You can search it, you can query it, but you can't take out a context and hand it to any AI. That last step, the distilling, it hasn't quite taken.

Understood, Not Just Counted

So how should distilling actually work, if it's going to produce a better personal context?

DDC's answer is that a real personal context is a file that can account for the traces of your actual life, your long-term preferences, your behavioral patterns, your judgments about what matters, and where you are right now. It's a file a person can read and an AI can use. Working from that, DDC has been building a lightweight distilling tool and named it Life Capsule. Distilling, the way Life Capsule defines it, means pulling the substance out of raw captures, dropping the noise, and organizing what's left into a Markdown file that a person can read and a machine can use. You pick the raw material yourself. It might be a screen recording of you scrolling TikTok, an order list from Amazon, a scroll through your Instagram feed, or your iOS Screen Time, your browser tabs, your desktop workspace, your calendar. All of it starts out messy, screenshots and recordings and lists and system screens piled together. After distilling, it turns into clear information. What you've been paying attention to lately, what you keep buying or saving, which tools you reach for and when, whether your interest in something runs long or was just an impulse, what you actually prefer when it comes to work, spending, learning, and travel.

This only became doable now because multimodal models finally got good enough. Turning recordings, screenshots, and system screens into high-quality text reliably used to be hard. Vision models today can read the structure, the content, the numbers, and the meaning on a screen, and the path from your life on screen to a personal memory opened up for the first time.

Using it takes three steps.

Capture. You pick a shortcut, Instagram, TikTok, Amazon, Google Maps, whichever fits, or you drag in images, upload a screen recording, upload a video of your desktop. One principle matters here. Life Capsule doesn't ask you to fill out a personality questionnaire or play a character. It only wants the digital traces you actually left, because what a person really watched, bought, searched for, and saved tends to be more accurate than what they write down under "things I like".

Review. A vision model reads your images or recordings frame by frame, then merges them into one coherent, readable draft. What it's after is what the recording actually says, not a UI description like "there's a button here, an icon there." Which content, numbers, products, posts, and settings are in there, and which of it could serve as context the next time you use an AI. At this step, you can look it all over and edit it, and confirm whether the AI really understood what you uploaded.

Life Capsule. Once you confirm, the Markdown goes into My Life Capsules. You can download it to Obsidian, or paste it into ChatGPT, Claude, or any AI, as system context or long-term memory. It keeps version history too. Every save is a revision, you can diff them, and watch how your interests, preferences, and state have shifted. It grows along with you.

Some people will worry that recordings and screenshots carry plenty of sensitive information, and that handing them to Life Capsule raises the same safety problem. In the words of DDC's cofounder, what we don't do matters more than what we can do.

Before you upload, you choose which part goes in, and you can mask anything yourself. After the vision model has read it frame by frame, you get to go over the draft that comes out, change whatever you don't want kept, and only once you confirm does it become a Life Capsule. From deciding what goes in to deciding what stays, every gate is yours.

Next, DDC will be integrating its privacy protection technology into Life Capsule. DDC's own work is infrastructure for AI data privacy and data assetization, and its product lineup includes an Ethereum Layer 2 built in house to carry data authorization, privacy proofs, and value settlement. Once that's wired into Life Capsule, your raw data still stays on your side, and what runs on chain is the authorization and the proof. Which AI, at what time, gets to read which part of you is yours to decide, and you can revoke it whenever.

In the end, the value of a personal context comes from being real, precise, and under your control, not from how much of it there is. Real, because what gets distilled is what you actually did, not a self-description you wrote on a form. Precise, because you distill what your AI needs. Want it to help you pick something out, upload your Amazon order list. Want it to be a partner at work, upload your desktop workspace, your browser tabs, your calendar. Under your control, because what goes in, what stays, and which AI gets it are yours to call from start to finish.

That's the Life Capsule DDC building. It wants AI to understand you, not just everyone.

Read more »

Everyone's Racing to Scale Models. DDC's Working on Something Else — Notes from ETH Summer Shanghai

The AI wave shows no sign of slowing, and Web3 keeps looking for ways to build alongside it.

Every summer, ETHPanda and LXDAO run "ETH Summer," a series of in-person Ethereum community events across multiple Chinese cities, bringing local Web3 builders together to talk through what's happening in the ecosystem right now.

On July 5, 2026, the event came to Shanghai, this time centered on AI and the economics of the AI ecosystem, drawing enthusiasts from across the space. DDC co-founder and CTO Li Yingxuan took part as both a speaker and a panelist, sharing our team's work and thinking on Web3 + AI from two angles, the product side and the data economy.

In this piece, we've pulled together the core ideas from that talk, for builders and AI enthusiasts who couldn't make it in person.

Keynote | AI's missing piece is you

During the keynote, our co-founder shared his research and views on "Training AI That Actually Knows You: How to Distill Your Personal Context".

For the past few years, almost every AI project has been pouring energy into the model, making it smarter with each iteration.

But when you actually use AI, you want it to get what you're saying and handle things in a way that fits you. A smarter model doesn't get you there. It needs data about you before it can understand you or do the job right.

There's a distinction here that's easy to miss. What big labs feed their models is massive general data, and what comes out is a general-purpose ability, built for everyone, a little bit of knowledge about everything. The problem with general data is that it can never fully land on you as an individual. So in day-to-day use, you still need to bring your own personal context, the part that's only about you.

That's what DDC works on.

Once AI big labs have handed us a smart assistant, the next step is that we need to bring our own personal context, so that assistant can grow into a real partner that's actually yours. But this is exactly what individual users are missing, because that data sits trapped inside different apps, with no easy way to pull it out into something readable.

So, at the event, DDC shared our new AI product Life Capsule, for the first time. This tool can help you take the real digital traces you leave scattered across platforms and distill them into a readable, portable Markdown memory file you can hand to any AI.

The whole thing turns on that word, distill. It pulls the substance out of messy raw traces, drops the noise, and shapes what's left into context a person can read and a machine can use. The raw material might be a screen recording of you scrolling TikTok, or an Amazon order. After distilling, it becomes clear information: what you've been paying attention to lately, what you keep buying, which tools you reach for. With this context in hand, AI doesn't have to guess who you are. It can actually read the traces you've really left behind.

To deliver on that "easy," DDC built Life Capsule as a very lightweight tool. Distilling takes just three steps:

• Capture — upload the real digital traces you've generated, catching your actual online behavior.
• Review — the vision model reads your uploaded images or screen recordings frame by frame, and you're free to look them over and fine-tune.
• Capsule — confirm the file and turn it into Markdown, ready to import into your notes app or an AI memory layer anytime.

There's no denying the big labs are working on memory and connections too, but this kind of approach has two built-in limits.

First, the distilling is unstable. The system pulls things out of your conversations automatically, and you have little control over what gets picked up or how accurate it is. Second, what it distills is only the version of you that shows up in the chat box. The much larger part, the things you never said to it but actually do day to day, stays invisible to it, so the picture is always partial.

Some people might worry that personal behavioral data reveals so much that handing it to Life Capsule raises safety concerns. In our co-founder's words, what a tool refuses to do matters more than what it can do.

For now, Life Capsule is set to read only the traces users upload themselves. It doesn't scrape anything without permission, and it doesn't use that data to label users or build any kind of profile or diagnosis. Data stays on the user's side by default, under their own control. Whether to export it, and which AI to hand it to, is entirely the user's call.

This way, users can distill their context with privacy intact, and use AI more intelligently as a result.

Panel | Data is an asset you own.

In the panel, the conversation moved from the product to something bigger, the question of how data should be used once AI agents actually start doing things on people's behalf. Below are a few of our co-founder's takes from the discussion. We'll lay out the fuller version in a follow-up piece.

First, for an agent to go from a smart tool to a real participant in the economy, everything hinges on whether it can get authorized access to genuine data. An agent that doesn't know the user, even with hundreds of tools at its disposal, can only give generic answers, and may even make the wrong calls. So if you want it to truly act for you, step one is understanding the user.

Second, understanding the user means using the user's data, and this is exactly where blockchain comes in. The chain needs to handle a few things at the base layer, authorization, proof, settlement, and revenue distribution. The raw data still stays on the user's side with privacy protected, while the record of who granted access, what scope the agent used, whether it went beyond its permissions, and who the resulting value should go to, all of that goes on-chain, where it can be verified and traced.

Third, the role data plays is shifting in the agent era. It used to be more like fuel for training models, with platforms collecting it, models consuming it, and users getting almost none of the upside. Once data becomes the context an agent uses to carry out tasks, it moves closer to an asset. Users can grant access on their own terms, deciding which agent gets it, for how long, within what scope, and at what price.

Following these three points, DDC's goal is to make personal context a data asset that can be authorized, verified, and settled. This is where DDC is headed longer term, beyond Life Capsule itself.

The labs have your assistant ready. The rest is on you.

Sooner or later, everyone's going to need a portable AI memory layer they actually own. It doesn't belong to any single platform, you can export it, review it, grant access to it, and pull that access back whenever you want.

Life Capsule is where this starts, distilling the digital traces scattered across your apps into a memory file you control. The further step is making that personal context something that can be authorized, verified, and settled, a real data asset that belongs to the user.

It's a road DDC hopes to walk with more developers, agent teams, and data partners. As for how personal context actually gets distilled and used, and how the AI data economy gets working, we'll get into it in the research and pieces to come. Stay tuned.

Read more »

The Second Brain Productivity Trap: You're Organizing, Not Using

Sort Obsidian's community plugins by all-time downloads and look at the top ten. Seven of them are there to make notes look better or sit in a neater arrangement. Excalidraw for drawing, Dataview and Kanban for building dashboards and boards, Advanced Tables for formatting, Style Settings and Iconize for theming and icons, Calendar for laying out your daily notes. The tools built for the actual reason you take a note, being able to pull it back and use it later, barely make the list. Omnisearch, the one that just helps you find things, sits down at fourteen.

That ranking is the whole problem in one screenshot. A note only has two jobs. You record something, and later you use it. But almost all the energy now goes to a third thing that feels like progress, making the vault cleaner and more impressive to look at.

Note-taking has quietly turned into A Product Trap.

Structure and clean design used to serve the using. You kept a tidy folder tree or a fast search so the whole thing stayed simple and immediately usable, so that when this piece needed a point from the one you wrote last week you could reach it, or the note your work depended on showed up in a plain index the moment you needed it. The structure was light, and it existed to be used. Somewhere that flipped. What people maintain now is elaborate, the kind of setup you have to study before you can run it and keep tending after that. Nested tag taxonomies, plugin stacks, a graph view, a dashboard wired together with query syntax. It looks impressive and it takes real work to learn and to hold together, and none of that work is the note getting used. The hours go into a vault that looks good for its own sake. A clean structure and a filled-in graph read like real work, so you close the laptop feeling like you built something. Line up what actually changed and it's thin. You didn't write another note. Not one more got pulled back out and used for anything. The interface got more elaborate and the two jobs a note exists for, getting recorded and getting used, sat still the whole time. That is the trap. It looks like progress on your notes while producing none.

Tiago Forte, who coined the whole second brain idea, has said organizing adds the least value of any step in his system and is mostly overhead. One writer deleted seven years of notes and described what she'd built as a mausoleum, something that started replacing her thinking instead of sharpening it. Another spent weeks building beautiful folder hierarchies that held twelve notes, while the messy Google Doc where he actually dumped things held hundreds and was the only part he used. The pattern people keep landing on is the same. The system looks finished. The notes still aren't used.

There's a decent map of why people slide into it, and it has little to do with discipline. In a set of experiments published in the Journal of Consumer Research, people repeatedly chose tasks they could finish over objectively more valuable ones they couldn't, acting as if completing something paid off on its own, apart from the actual reward. A separate line of work named precrastination found people will rush to clear a small task even when it costs extra effort, mostly to get it out of their head. The pull is toward whatever closes fast and returns a result you can see right now. Open-ended work with a distant, uncertain payoff gets pushed down the pile.

So the newest work on notes is circling back to those two jobs.

The using side is already moving. The AI now landing inside note apps is aimed squarely at retrieval, at getting an old note back to you when it matters instead of leaving it to sink. Obsidian's Copilot, past a hundred thousand users, lets you ask your whole vault a question and pulls the relevant notes into the answer. Smart Connections surfaces related notes beside whatever you're writing. The newest builds go further and run agents like Claude straight inside your vault. And it's started running the other way too, with connectors that let an outside assistant like Claude or ChatGPT reach into your notes directly, so the vault feeds the model you already use instead of housing its own.

And, once an AI answers from your notes, whether from inside the app or through a connector, your vault stops being something you read and becomes something the model reads, a database it queries. That shift is already an industry of its own. Over the past couple of years a whole category has formed around turning a person's scattered information into a memory layer an AI reads from, with tools like Mem0, Zep, and Letta treating that store as its own component sitting between you and the model, pulling the relevant pieces into context the moment they're needed. Mem0's engine distills the raw material down to what matters instead of loading all of it in. By 2026, this memory is treated as a first-class part of the stack, with its own benchmarks and research, no longer chat history crammed into a prompt. Your notes are turning into the database an AI runs on.

Recording needs to move too. The original way is manual, you type things in and organize them yourself. That suits an old-school note-taker, someone who likes doing their own excerpting and wants the act of writing it down to be part of the thinking. But once the goal is for the vault to serve as your context store, manual input stops being enough. You can't hand-type your way to a full picture of yourself, and most of what matters never gets written down in the first place.

The better way is to distill. You take the raw material, a document, a session, a stretch of your own activity, and compress it into a clean markdown summary, then paste that into a note or drop the file straight into the vault.

For a document you read, you can do that yourself. For what you actually do all day, you can't. The record of which tools you lived in, what you searched, what you kept coming back to, is too large and too continuous to hand-distill, and it's exactly the part that never gets written down.

Today, capturing what you actually do and turning it into context an AI can use is already a funded race. Rewind ran in the background on your Mac and indexed everything that crossed your screen, then became Limitless and got bought by Meta at the end of 2025. Microsoft shipped its own version in Recall. Screenpipe does it in the open, fully local. The approach has been shifting, though. The early tools stored raw screenshots, which was heavy to keep and uncomfortable to run, and part of why they struggled. The newer ones read the activity and keep only the text. Littlebird raised eleven million dollars in early 2026 on that pitch, reading your screen and storing the context as lightweight text instead of images. DataDance sits on the same edge of the field with Life Capsule, taking a screen recording of what you did and distilling it into a clean markdown profile.

That text-first move is the one that fits a vault. A markdown profile isn't a screenshot you have to house and protect. It's the same file type your notes already are, so it drops into the store beside everything the AI reads, or straight into Claude and ChatGPT. The recording gets turned into a short block of context rather than a video no one replays. And the recording side of your notes finally gets the input it was always missing, the version of you that shows up in what you do, not only in what you sat down to type.

That is the whole shift. The vault was a productivity trap, a place you kept busy organizing, tidying, and arranging, work that always looked like progress and rarely turned into any. What it's becoming is the opposite kind of thing, a store that holds the real record of what you do, distilled into something an AI can read and act on. One you kept for its own sake. The other gets used.

The title of the trap was Organizing. What replaces it is Using. You don't need a cleaner place to keep your notes. You need a place your notes finally do something.

Read more »

Why AI Still Doesn't Know You, No Matter How Much Memory You Give It

Your AI knows a lot about you now. It knows you are a developer, that you prefer Postgres, that you like things explained without padding, that you have been chipping away at a side project you mentioned a few weeks back. Ask your AI about yourself and it can list these things back to you without missing a beat.

And it knows them because you taught it. Early on it was small stuff, a correction here, a preference there, the kind of thing memory quietly picks up. Then you went further. You wrote out proper custom instructions. You set up a system prompt you were actually happy with. Maybe you connected your own notes and files, or wired up retrieval so your AI could pull them from your own material. By any reasonable measure, you did the work.

So here is the honest question. When you ask AI to do something that matters, create in your voice, make the call you would make, draft the thing the way you would draft it, are you satisfied with what comes back?

And if you are being honest, the answer is NO.

If you have gone deep with your AI, maybe you only tweak a few parts and it lands close to what you wanted. But more often, the answer it gives you gets thrown out and redone, again and again. Sometimes it broke a rule you set. Sometimes it reads too generic. Sometimes it just is not what you meant. In the end, you keep a piece of it, maybe, and create the rest yourself.

So after all that material and all that back and forth, why can it still fall short of fitting you? That gap is what this article is about, and the answer is not the one most people reach for, which is usually to go feed it even more.

Why feeding your AI more doesn't make it know you

Feeding your AI more material makes sense. The stuff you give it is precise and it is yours. It is not something the model made up, and it is not the generic average sitting inside the LLM already. Your own profile, your past work, the way you described how you think, none of that is reliable in the pile of training data the model came with. You are handing it the specific you. And the more you hand over, the more dimensions, the finer the detail, the better it should know you.

But look at what all of that material actually is. Your custom instructions are you describing yourself. Your memory is things you told it, or things it picked up from what you told it. Your notes, your files, your retrieval setup, all of it reaches into material you wrote down and handed over. Every piece of it is the same kind of thing. It is the account you give of yourself.

And that account, however detailed, is partial. It is one side of you, the side you can put into words and chose to. And there is a reason that side does not add up to the whole. What people say about themselves and what they actually do are not the same thing, and the gap is not small. When researchers compare the two directly, asking people to state their preferences versus watching what they really choose, the behavior wins easily. In one study from a team at Warwick, models built on what people actually did predicted real-world outcomes about 1.5 times better than the best model built on what they said, explaining 49 percent of the variance in actual sales against 32 percent for the stated version.

You can feel why in something as ordinary as dating. At speed-dating events, people report caring most about personality, then turn around and choose based on looks. The stated answer and the revealed one come apart, in the same person, minutes apart. It is not that anyone is lying. People are often genuinely unsure what drives their own choices, their preferences shift with context, and most of us describe ourselves a little closer to who we would like to be. So the profile you give your AI, however honest you are, is the version of you that you can put into words and want to present. The fuller, less flattering truth lives in what you actually do.

That is the half your AI never sees. It is sitting in what you do all day and never think to write down. The things you reached for instead of the things you said you wanted. The books you actually finished. What you go back to, what you abandon, what you pay for, what you keep open at one in the morning when no one is watching. None of it makes it into a profile, and it is the most honest record of how you actually operate.

It has another advantage too. It keeps up with you. What you write down is fixed the moment you write it. You set up your profile once and it sits there, but you do not sit there. Your tastes move, your situation shifts, what you cared about three months ago is not quite what you care about now. A profile is a photo of who you were the day you wrote it, and you keep being asked to recognize yourself in a picture that stopped updating. Your behavior never freezes like that. It is produced fresh every day, on its own, whether or not you ever sit down to describe it.

None of this means what you say about yourself is worthless. For some things a direct answer really is the best source, and there are traits people report about themselves quite reliably. Stated information is not wrong. The trouble is that knowing someone takes both sides, and right now your AI runs on one of them. The whole apparatus of personalization, the memory, the instructions, the retrieval, sits on the side you can describe. The other side, the bigger and more honest one, is almost entirely absent.

And that is the real reason your AI still does not know you. It is not short on detail about the you that you write down. It barely has any of the you that you live out.

Which part of you would actually help your AI

It helps to get specific about what that half actually contains, because different things you ask of AI need different pieces of it.

Take creative work, where people complain the most about AI sounding generic. You tell it to write in your voice, and it has your instructions, maybe a few samples you pasted in. What it does not have is your taste, and your taste lives in your behavior. The articles you read to the end and the ones you closed after a paragraph. The sentences you highlighted. The drafts you wrote and deleted. The posts you saved and went back to. That record is what actually separates your voice from a competent average, and none of it is in the prompt. You described your style in a sentence. Your style is sitting in a thousand small choices you never articulated.

Now take decisions, like shopping or planning. You ask AI to recommend something and it leans on what you told it you want, which is exactly where stated and real come apart. What would actually help is the record of how you decide. What you compared before buying, what you almost bought and backed out of, what you keep repurchasing, what you returned. The price you talk about caring versus the price you actually pay. That pattern predicts your next choice far better than the preferences you would list if someone asked.

It runs through everything else too. How you really plan a week versus how you say you do. Which topics you return to on your own time. What you start and quit, what you finish. In every one of these, the useful signal is not the summary you can give of yourself. It is the behavior underneath it, and that is the exact part your AI is working without.

All of which would be a smaller problem if this information were hard to come by. It is not. You produce it constantly. It just does not end up anywhere your AI can use it.

The half you don't yet hold

The industry already senses this, which is why everyone is racing to connect AI to more of your life. ChatGPT can now plug into Google Drive, Gmail, Slack, and over a hundred other apps. Gemini's Personal Intelligence reaches into your Gmail, Photos, YouTube, and search history, and uses them to recommend a book or plan a trip around where you have actually been. The bet behind all of it is the one this article has been making. What you type is not enough, so go reach for what you actually do.

But look at how far it reaches, and where it stops. ChatGPT's connectors search one app at a time, and you have to know which app holds the answer before you ask. Gemini only sees the slice of your life that happens inside Google. The shopping you do elsewhere, the things you read in another app, the long trail you leave across the dozens of services that are not Google, none of it is in the picture. Each tool reaches into its own corner and calls it context.

And that is the real shape of the problem. The most honest record of how you operate is not missing because it is private or rare. It exists, in enormous detail, and it is being added to every day. It is just sitting in a hundred separate places, each one holding its own fragment, none of them yours, none of them talking to each other, none of them reaching the AI you actually work with.

So the next time your AI hands you something competent and faceless, be precise about what went wrong. You did not under-explain yourself. You explained the part of you that can be explained, and that part has a ceiling. The half that would have made the answer feel like yours was never on the table, because it has never been in your hands to give.

Until it is, more memory will only sharpen the picture of the person you can already describe. The one underneath, the one you actually are, stays a stranger.

Read more »

Your Second Brain Is One Piece of Jarvis. Here's the Rest.

When the first Iron Man hit theaters in 2008, nobody expected a voice to become a breakout thing. Jarvis, short for Just A Rather Very Intelligent System. It runs Tony Stark's house, manages his work, and knows all his needs.

And, it did not stay on screen. For over a decade, people have been trying to build their own Jarvis, with open-source projects named after it, weekend builds running on a Raspberry Pi, paid courses promising to make you Tony Stark. The dream kept pulling people in.

The most visible attempt came from Mark Zuckerberg. In 2016 he made building his own Jarvis his challenge for the year, with resources almost nobody else had, the full stack of Facebook's AI tools across language, speech, and face recognition. He even got Morgan Freeman to voice it.

After about a hundred hours, Zuckerberg's Jarvis could control the lights, the temperature, and the music, open the gate by recognizing faces, and wake his daughter with Mandarin lessons. The piece he was proudest of was small but telling. It learned his and Priscilla's taste in music, so when she asked for a song, it leaned toward her preferences over his.

Everyone else's attempts landed in roughly the same place. A lot of effort went into wiring devices together and getting the system to understand speech, and what came out could hear a command and carry it out. Turn on the lights. Play some music. Look something up.Which, if you think about it, is just the smart home you already have. Ask Alexa for the weather, tell Google to turn off the lights, have Siri set a timer. These do exactly what they are told, and nothing past that. They wait for a command, match it, and execute. While Jarvis never waited. It understood the situation and acted on its own, which is exactly what none of these can do.

Zuckerberg said it himself. Everything he built that year was a variant of the same basic pattern recognition, and even a thousand more hours would not have produced a system that could learn new things on its own. AI, in his words, was both closer and further away than people thought.

But that verdict came with an expiration date. The technology of 2016 and the technology of today are barely the same thing, especially with everything that has happened since in AI, and now AI agents. Looking back, each leap was quietly supplying Jarvis with another one of the core capabilities it would eventually need.

The brain: from following commands to thinking on its own

What made Jarvis special was that it had a personality and could think on its own. It did not wait to be told exactly what to do. In the harbor battle at the end of Iron Man 3, Tony gives one command, the House Party Protocol, and Jarvis does the rest, calling in dozens of suits and flying the whole Iron Legion into the fight on its own.

But almost every attempt before now worked the other way, matching an instruction or a keyword to a fixed response. So before anything else, building a Jarvis means building a brain.

Large language models gave Jarvis that. A model can read what you say in plain language, follow what you mean, reason through it, and answer in a way that holds together.

Understanding still waits for instructions, though. The next piece was getting AI to carry out a task without being walked through every step. Automated workflows did part of this, chaining actions so a single trigger could run a whole sequence on its own. But a fixed workflow only does what it was set up to do, and the moment something unexpected shows up, it stalls.

AI agents closed that gap. An agent takes a goal, breaks it into steps itself, picks the tools it needs, does the work, checks the result, and adjusts. You do not hand it a command and wait. You hand it an outcome and it works out how to reach it. A chatbot answers your question. An agent goes and does the thing.

There is also the matter of personality. Part of what made Jarvis feel like Jarvis was that it had a consistent character, a dry wit, a way of speaking that was unmistakably its own, rather than a blank voice reading answers. That used to sound like the soft, unbuildable part of the idea. It is now an actual field of research. Anthropic has worked on what it calls persona vectors, patterns inside a model that correspond to specific traits and can be monitored and steered. Other groups are using established personality frameworks like the Big Five to define and hold a stable character across long stretches of conversation, so a system can stay recognizably the same instead of resetting to a generic tone every session. The work is early, but the direction is clear. A consistent personality is no longer off the table.

Put those together and the inside of Jarvis starts to exist.

The record: when your life started leaving a trail

A brain is only useful if it has something to work with. Jarvis knew Tony because it could see everything Tony did, where he went, what he built, how he worked, who he let in. A mind with nothing to go on is just a clever stranger. So the second piece is the material, the record of a life for the brain to draw on.

For most of history that record did not exist. What you did simply happened and then was gone. You read a book and the reading left no trace. You walked into a shop, picked something up, changed your mind, walked out, and none of it was written down anywhere. Your days were full of choices that disappeared the moment you made them. There was nothing to look back on.

That is the part that quietly changed. Once your life moved onto screens and through the internet, your behavior stopped vanishing and started being recorded. Now the book you read is an ebook that knows which pages you lingered on. The shop is an app that logs what you viewed, what you added, what you dropped before checkout. What you watched, what you skipped, what you replayed, what you searched at two in the morning, all of it is captured somewhere. The same actions that used to evaporate now leave a trail of data behind them.

It is easy to miss how big a shift that is. For the first time, the raw record of how a person actually lives exists in a form a machine can read. Not a description of you, not what you said about yourself, but the actual log of what you did. That is exactly the kind of material a brain would need if it were ever going to know you the way Jarvis knew Tony.

The knowledge: when people started saving their own minds

A brain that knows you still needs something to know.

Jarvis was not just good at understanding Tony. It also held an enormous amount of knowledge and could pull up whatever Tony needed the moment he needed it. Ask it to analyze a material, dig up an old file, run the numbers on a design, and the answer comes back right away.

So the third piece is a body of knowledge for it to draw on, and ideally, your knowledge, the things you have gathered and want to keep.

This is where the second brain comes in. The instinct to keep your own knowledge somewhere outside your head is old. People kept commonplace books for centuries, copying down quotes and passages they did not want to lose. What changed is the form. The notebook in a drawer became a digital system, searchable, linked, built to be returned to instead of forgotten.

Over the past several years that habit turned into a movement. A huge number of people took it up under the same name, the second brain, and it spread far. Notion crossed a hundred million users. Obsidian grew to over a million people building their own linked webs of notes by hand. The book that popularized the term sold half a million copies in dozens of languages. A whole genre of videos and templates grew up around teaching people how to do it well.

What all these people are doing, without necessarily putting it this way, is moving a part of their own mind out into the open. The articles they thought were worth saving, the ideas they wrote down so they would not slip away, the connections they drew, the conclusions they reached. It is the thinking they cared about enough to set down and keep.

And that is exactly the kind of material that can work for Jarvis. It is the knowledge you have already judged to be worth having, gathered in one place. For the brain we have been describing, this is gold. It is not random information off the internet, it is your reference library, the things you have read, decided on, and chosen to hold onto.

The piece still missing

So the three pieces of Jarvis are finally real. A brain that can think and act. A record of how you actually live. A store of your own knowledge. Put them together and you have the thing people have been chasing since 2008.

The brain is arriving on its own, pushed by the whole AI industry. Your knowledge you already keep, in your second brain. The piece that is still stuck is your behavior. It gets recorded all day long, but it sits locked inside the apps that capture it, scattered across a hundred silos that have no reason to hand it back to you.

That is what is missing. Not a smarter model or a better notes app, but a way to get your own behavior out of those silos and join it to the rest.

Read more »

How to Make Your Second Brain Actually Work With AI

Do you think your second brain is actually working? Look at how you treat it.

Maybe you still tidy it up every day.

Maybe you kept it going for a while and then drifted off.

Or maybe you genuinely use it, pulling notes back out when work calls for them.

But either way, the real question is the same. Even with the whole system built, does it feel anything close to what you pictured when you started?

A second brain was always meant to be used, not just filled. People take notes so they can come back to them later instead of keeping it all in their head, which is the whole point of a second brain. The catch is that "later" rarely came.You captured far more than you ever pulled back out.

AI has started to fix the first half of that problem. It can dig through your library and pull up what you need without you combing the shelves by hand. But finding a note was never the whole point. Getting it to actually work for you is. So what does that next step look like, and how far can a second brain really go right now? A few approaches are already taking shape. That is what we are getting into here.

How most people still use their second brain today

Start with how this actually works day to day. You write things down, save them, and leave them alone until some project pulls you back. One day you are working on a topic and it hits you that you probably wrote something about this before. So you go looking for it.

That looking-for-it part used to be slow. Before tools like Obsidian and Notion, you went through your notes one by one, opening things, scanning, trying to remember where you put it. The newer tools fixed a lot of that. Obsidian builds a web of linked notes, so related ones surface on their own. Notion can search across your whole workspace and answer questions about what is in it. Some newer apps like Mem go further and use AI to resurface notes you forgot you had. Finding things is genuinely easier now.

But look closely at what actually improved. It is the finding. Every one of these makes it faster to locate a note, and then leaves you exactly where you were before, holding a note you still have to do something with. You read through what came up, pick the parts that matter, copy them, and paste them into Claude, ChatGPT, Gemini, wherever the real work happens.

So the tools made the search faster. They did not change what happened after the search. You are still the one who has to remember the note exists, judge what is worth keeping, and carry it into the AI by hand.

Your second brain is not doing the work here. You are. You are the one carrying notes into the AI by hand.

How some people are starting to make second brains work with AI

But, used the old way, these tools never come close to their potential. The note apps we call our second brain can do a lot more now that AI is in the mix.

And a small group of users has already started digging into what that potential actually looks like. Here are the three main ways they are doing it.

Path 1: AI built into the note tool itself

The most direct approach is to use a tool that ships with AI inside it.

Notion is the most familiar example. You write and store everything in the same place, and the AI lives right there with it. Ask what you concluded about a client last quarter and it reads across your pages to answer. Tell it to turn three weeks of meeting notes into a project brief and it pulls from what is already in your workspace.

Tana works the same way, except it gets you to label things as you write so the AI has cleaner material to work with. Every meeting note gets tagged as a meeting, every client note as a client. You are quietly building a structured library while you take notes. So later you can ask Tana to pull every action item still open across all those meetings, and it answers from the structure you have been building, instead of making you dig through the notes one by one.

What this really removes is the trip. In the old way, your notes sat in one place and the AI lived in another, and you were the courier between them. Here there is no courier. The material and the intelligence share an address. For a lot of everyday work, summarizing, drafting, pulling threads together, that alone is a real jump. The second brain stops being a drawer you open and starts being a desk you work at.

But this path has a wall around it, and the wall is the tool itself. Everything works beautifully as long as you stay inside Notion, or inside Tana. The moment you want to take that thinking somewhere else, into Claude, into a different agent, into whatever tool fits the job better tomorrow, you are back to being the courier. You copy things out, you paste them in, you re-explain the context the new tool has never seen. The second brain works, but only at one address. Step outside and you are carrying notes by hand again.

Path 2: Installing AI into your notes

The second route is the open-notes approach, and it is the one getting the most attention in the AI crowd right now. Obsidian is the main example. It works as a local markdown editor, so everything you write gets saved as plain .md files on your own machine. That format matters, because markdown happens to be the format AI tools read most easily, with no conversion needed.

On its own, Obsidian has no AI in it. The AI comes from plugins, and now the community has built a lot of them. They sit at different levels.

The simplest plugins are about retrieval. Smart Connections is the common starting point. It builds a local index of your vault and, as you write, surfaces older notes related to what you are working on, even ones that share no keywords with it. Write about remote team communication and it can pull up a forgotten note on async video updates. It runs locally, so nothing leaves your machine. This is genuinely useful, but it is still helping you find things rather than do anything with them.

The next level of plugin starts writing with your notes, not just locating them. Text Generator drafts, rewrites, and expands text inside the editor, turning your bullet points into prose without leaving Obsidian. Copilot for Obsidian goes further, adding a chat panel that answers from your actual vault. Ask what you concluded about pricing last quarter and it replies from your own notes instead of making you dig them up and paste them into a separate chatbot. With its agent mode, you can hand it a larger task and let it work across many notes at once. At this point, you have stopped searching and copying. The AI is producing things directly from what you already wrote.

The furthest level hands your whole vault to a full agent. Khoj can run as your own self-hosted assistant across your notes and even build custom agents over them. And there is a community plugin that connects Obsidian to Claude through MCP, which lets Claude Code or Claude Desktop read and write your notes directly and treat your vault as the folder it works inside. You can ask Claude to go through your notes, connect ideas across them, and write something back into the vault, and it works against your real files rather than a copy you pasted somewhere. The notes finally get used without you doing the carrying.

The catch with this whole route is that you have to build it yourself. You pick the plugins, connect the model or the agent, handle the setup, and fix it when an update breaks something. How capable your second brain gets depends on how well you put it together. For people who like building their own tools, that is part of the appeal. For everyone else, the setup and upkeep never really stop.

Path 3: Pulling your notes into AI

The first two paths both keep the work inside the notes. You are still in Obsidian or Notion, and the AI comes to you there. But a lot of people now want another thing. Some of them already live in one AI, Claude or ChatGPT, and just want their second brain available inside it. So the third path is about bringing your notes into the AI side.

Besides uploading your notes manually, a more lasting setup connects the AI to where your notes actually live, so you stop re-uploading. With MCP, you can give Claude a standing line into Notion, and after that you can ask it about something from the last quarter and it goes and reads the relevant pages itself. Some people connect a tool like NotebookLM to Claude, so one side holds all the documents and the other does the thinking, and you just talk to it. The notes stay in their home, and the AI goes and gets what it needs when you ask.

A few people take it one step further and leave the AI a kind of map. When they have Claude working over a folder of notes, they drop in a short file that explains how everything is laid out, what goes where, what the tags mean. Now every time they come back, Claude already knows its way around and they never have to re-explain the setup. It is a small thing, but it hints at where this is all heading, an AI that holds a standing sense of your notes instead of meeting them cold every time.

What you get out of all this is freedom of movement. Your notes are not locked inside one app's AI. You can take them to whichever model is best for the job, and hook the same notes up to more than one.

Why a working second brain still needs more than your notes

So all three paths get your second brain finally callable. The AI can reach into it, read it, build on it. But sit with what comes back for a minute and something feels off. It is competent enough. You asked it to write the way you would, and it tried. It picked up a bit of your phrasing, leaned on the notes you pointed it at. But what you get reads like an AI doing an impression of you, close in places and hollow in others. Half the time it does not even hold the specific things you asked for at the start.

Here is why. Your second brain handed the AI everything it had, and everything it had was notes. A second brain, as most people have built it, is a pile of things you decided were worth writing down. That pile does not carry how you actually think, the phrasing you reach for, the images you pick, the way your mind moves from one idea to the next. Give a model only that and it fills the gaps with what a generic competent person would do. The result is part you and part the average of everyone.

You might say the AI tools already handle this, and it is true that they have started to. Each of the big ones builds a picture of you, just in a different way, and the differences say a lot about how far this idea has come. ChatGPT leans on what you type in yourself, a profile and a set of instructions you write by hand, plus a memory that saves facts as they come up. Claude leans the other way and does it on its own, quietly distilling your conversations every day into a running sense of your work, your tone, the tools you keep using. Gemini goes further, reaching past the chat window into your Google world, your searches, your YouTube, the things you actually did rather than the things you told it.

And it works, up to a point. Each step makes the AI feel a little more like it knows you, and the replies come back fitting a little better than they used to. But the picture each one builds is only as wide as what it can see. ChatGPT knows the you that fills out forms. Claude knows you who talks to Claude. Even Gemini, reaching the furthest, only knows the you that lives inside Google. Personalization is real, and it is also partial, drawn from one corner of your life and treated as the whole of you.

Still, look at the direction all of this is moving. The trend line is hard to miss. The closer anyone tries to get to a second brain that truly knows you, the more they lean on behavior instead of notes.

That is the tell. The missing piece in a second brain was never more notes or better notes. It was the part of you that notes were never going to capture, the real record of how you choose, sitting in what you do rather than what you write down. The whole industry is inching toward it from different sides. None of them has the whole of it yet, because the whole of it does not live inside any single app. It is spread across everything you use.

What it actually takes for a second brain to work

So it is worth saying plainly what working really means here again, after everything above.

In the last article, we came back to what a second brain was always supposed to be. Not a filing cabinet, but something closer to a partner that thinks with you, the kind of thing Tiago Forte described long before the tools could deliver it.

FYI: https://datadancewallet.blogspot.com/2026/06/is-your-second-brain-just-knowledge.html

Hold a second brain to that definition and the test gets simple. It is not the one with the most notes or the fastest search. Those make what you wrote easier to reach, and reaching what you wrote was never the point. A second brain works when it can act on who you actually are, and that takes more than half of yourself when you sat down and record. How you really operate, what you choose, what you return to, what you do when no one is watching, lives in your behavior, the more honest half that notes never catch.

The move still ahead is from connected notes to connected context. From a second brain that remembers what you wrote, to one that knows how you actually think, choose, and work.

Read more »