AI Is Quietly Taking Your Memory
AI companies are racing to own users' context and memory through integrated tools like Claude in Slack and ChatGPT's auto-memory features, creating a lock-in trap. The speaker argues users should own their context (expensive, irreplaceable) in local plain text files while renting the AI model (cheap, replaceable), and demonstrates his personal knowledge assistant system built on markdown files that work with any AI model.
Summary
The speaker outlines a critical industry pattern where AI companies are simultaneously launching tools designed to capture and retain user context and memory. Claude Tag (launched June 23rd) integrates into Slack as a persistent teammate, while ChatGPT (June 4th) introduced auto-memory that builds profiles of users across conversations without manual curation. Andrew Karpathy called this the third major redesign of AI interaction, after websites and apps—now AI agents are self-contained entities living within organizational tools.
The speaker identifies the underlying incentive: since AI models have become commodities with better versions shipping every few weeks, companies can no longer lock users in through the model itself. Instead, they're racing to own the context layer—the decisions, history, relationships, and operational memory that make each organization unique. This creates a trap that feels like pure convenience on day one (a tool saves 20 minutes, remembers team decisions, learns shorthand) but becomes dangerous over six months when users realize their entire operational memory and decision-making context lives inside a vendor's system. The speaker uses the example of note-taking apps (Roam, Obsidian, Notion, Mem, Tana) where shutdowns or pivots meant losing not just notes but the thinking structure built within them.
The critical distinction is that the AI model is the cheap, rentable part (billions to train, but costs users only token-by-token, constantly improving, impossible to own), while context is the expensive, irreplaceable part (years to build, specific to each organization, impossible to recreate if lost). Most users are inadvertently trading away their expensive context to rent the cheap model—exactly backwards.
The solution is to own the context layer in your own infrastructure and let AI models compete to be the brain running on top. The speaker demonstrates his own system: a local folder of plain text markdown files containing projects, people, decisions, and cross-connected information, with AI agents running as contracts (also plain text) on top. This architecture is inspectable (readable plain text), portable (just files on a disk), and model-neutral (can swap Claude, ChatGPT, Codex, or Gemini without changing anything). He deliberately turns off auto-memory features in both Claude and ChatGPT, controlling what gets saved in his folder based on his own rules rather than the AI's decisions.
The speaker calls his system myPKA (personal knowledge assistant) and offers the entire scaffold freely. He notes he uses the command-line interface rather than the pretty chat app specifically for more control—control over what the agent reads, which model runs, and what can touch the folder. While this seems technical, he argues it's not intimidating for non-coders; it's just typing what you want and having it done.
Key Insights
- Every major AI company (Claude, ChatGPT, Codex, Gemini) is simultaneously launching persistent context-capture tools because models have become commoditized—companies can no longer lock users in through the model itself, so they're racing to own the memory and context layer instead
- The trap springs after three to six months when users have stopped writing things down anywhere else because the AI already knows everything, and only then do they realize their operational memory and decision-making context lives entirely within one vendor's system
- The AI model is the cheap, replaceable part that users are renting by the token while giving away the expensive part—their context—for free; this is exactly inverted from how it should be
- Clem Delangue at Hugging Face stated that the peak of proprietary API lock-in has been reached and the industry is moving toward a more balanced world where open models and owned infrastructure will get a much bigger share
- A model-neutral architecture where context lives in plain text files with AI agents as contracts allows instant model-switching without losing any operational memory—demonstrated by swapping Claude to ChatGPT to Gemini while pointing at the same folder
Topics
Transcript
[0:00] if you follow me on this channel, you know my whole life and business runs through this one local folder. plain text files on my own disk, and on top of it, I run a whole team of AI agents. Right now, Claude is the engine. Now look at this. I swap Claude with Codex. Same folder, same agents, same contracts. I just swapped the brain on top, and I could d-do the same again to Gemini, GLM, and whatever you like to use. Now, here's why I'm showing you this today. Two days ago, Anthropic put Claude inside Slack, a team communication tool. [0:32] so look at what they actually said. a new way for teams to work…
Full transcript available for MurmurCast members
Sign Up to AccessMore from ICOR with Tom | AI Productivity
I Can Build a $11 Billion Productivity App with AI (You Can Too)
The speaker demonstrates how modern AI models (Claude Sonnet 3.5) enable anyone to build sophisticated productivity applications without extensive coding knowledge, using the newly rebuilt myICOR membership platform as proof. He argues that pre-built platforms like Notion, Circle, and Mighty Networks are becoming obsolete as AI democratizes custom application development, and announces a live workshop teaching the conceptual and technical skills needed to build such apps.
I gave Claude 4 years of our work. It built this app.
A developer used Claude Sonnet 5.5 to build a fully functional ICOR-based productivity application in just one hour by providing the AI with four years of accumulated productivity knowledge and content. The resulting app successfully implements complex ICOR methodology concepts including inbox management, task prioritization, interruption handling, and life dimension tracking, though it needs UI/design refinement.
Claude just replaced every productivity app I use.
A content creator demonstrates how Claude's Sonnet 3.5 model with Code Interpreter can replace multiple productivity apps by building three fully-functional applications in under an hour with minimal prompting. The creator showcases a note-taking app with database features, a Mac application integrating their custom Obsidian plugins and ICOR life management methodology, and highlights the unprecedented speed and polish achievable without traditional development skills.
My AI team took 8 hours. Claude Sonnet 5.5 took 15 minutes.
A creator comparing an 8-hour AI team project with a 15-minute Claude Sonnet 5.5 solution reveals that excessive guardrails and restrictions paradoxically hindered the multi-agent system's performance. The speaker demonstrates how reducing constraints and context allows AI to produce more creative and functional results, though both approaches have significant limitations for production use.
Your knowledge should outlive every AI tool you use.
The speaker clarifies the distinction between the ICOR methodology (a tool-agnostic productivity framework) and its implementations like the ICOR for Life scaffold and myPKA AI team, emphasizing that users can adopt the core principles in any tool they prefer. They announce plans to restructure their membership platform by separating these components to reduce confusion and help users understand they're not locked into specific tools or systems.