Claude Cowork Analyzed 10 Years of My Data (It Knows More Than I Do)
A content creator grants Claude AI access to 10 years of personal data across multiple platforms and tools, allowing it to generate a comprehensive biography and evolution timeline from 2018-2026. The AI successfully synthesizes information from Gmail, Google Drive, Readwise, LinkedIn, and other sources to reveal the speaker's professional journey from corporate biochemist to productivity movement founder, while also uncovering unexpected privacy vulnerabilities in the process.
Summary
The speaker describes an experiment where they provide Claude AI (Opus 4.6) with access to their personal data ecosystem through multiple connectors (MCPs) including Gmail, Google Drive, Readwise, Todoist, ClickUp, and external web sources. Using a single prompt asking the AI to explain who they are and how they evolved from 2018-2026, the AI conducts an autonomous investigation across all available data sources.
The AI's discovery process reveals extensive information gathering capabilities: it accesses the speaker's Google Drive documents, reads their Readwise highlights and quotes from Kindle, reviews project management tools, examines email profiles, and even conducts external internet searches finding old blog posts, YouTube interviews from 2018, and public profiles on platforms like Clay and Paperlike that the speaker had forgotten about or wasn't aware existed.
The resulting biography synthesizes a detailed career narrative: PhD in biochemistry (2013), 8 years at Hoffmann-La Roche where they improved team performance by 60%, transitioning to YouTube content creation (2017), launching a membership program (2019), leaving corporate work (2021), founding the Paperless Movement (rebranded to My AI Core), meeting co-founder Paco (2022), and scaling through courses and proprietary methodologies. The AI identifies the speaker's consistent philosophy: productivity is about processes and systems, not tools.
The speaker notes the AI's minor misinterpretation (Project Phoenix as a real product rather than a demo) but emphasizes the overall accuracy and depth of analysis. They highlight the potential to expand this biographical analysis further by connecting personal journals from Day One (journaling since 2015) to include private insights alongside professional achievements. The speaker reflects on how this AI-driven biographical analysis would have previously required hiring researchers or biographers, positioning this capability as emblematic of a new era in AI-assisted personal knowledge synthesis.
Key Insights
- Claude AI was able to autonomously aggregate and synthesize 10 years of personal data across multiple disconnected platforms (email, cloud storage, note-taking apps, project management tools, and external websites) from a single prompt without explicit instruction on which sources to access.
- The speaker discovered unexpected privacy vulnerabilities through the AI's investigation, including public profiles and old blog posts they had forgotten about or were unaware existed, revealing that personal information is distributed across the internet in ways individuals may not realize.
- The AI identified a consistent underlying philosophy driving the speaker's 8-year evolution across different roles and ventures: that productivity is fundamentally about processes, clarity, and building systems that serve life rather than consuming it.
- The speaker's professional trajectory followed a specific arc from academic researcher to burnt-out corporate team leader to YouTube creator to founder of a global movement with proprietary methodology, app, book, courses, and team—synthesized by the AI from fragmented data sources.
- The speaker views this AI capability to generate comprehensive biographies as marking the beginning of a new era, where tasks that previously required hiring researchers or biographers can now be accomplished through AI analysis of connected personal data systems.
Topics
Transcript
[0:00] AI is able to tell me in seconds more about me, who I am, and what I have done in the past 8 years than I would ever be able to even gather within a week to write the story of my life. And all I did is giving it access to all my folders on my Mac, but also the connectors to my external tools such as Gmail, G Drive, Readwise, and so on. And yeah, you have to be fine with the feeling of that a system is going in there and sends all the data across the [0:31] world. But guys, curiosity just won and I did it, and this is the only prompt that I…
Full transcript available for MurmurCast members
Sign Up to AccessMore from ICOR with Tom | AI Productivity
Claude sucks. And here is why.
This video explores why Claude and other AI models produce inconsistent results, examining how model selection (Haiku vs Sonnet vs Opus vs Claude 3.5 Fable), effort levels, prompt clarity, and system context dramatically affect output quality. The speaker demonstrates these differences through a Venn diagram creation task and advocates for using organized folder systems with documented SOPs and code-based solutions to ensure consistent AI performance.
Claude can now prompt its own SESSIONS. I built the boss.
Claude now has the ability to interact with multiple sessions, allowing a streamlined orchestration of tasks through an agent called Larry. This integration can simplify task management and enhance productivity by enabling a single main session to oversee various sub-sessions or tasks.
I switched off Claude's memory. My folder does it better.
Tom explains why he disabled Claude's native memory feature and instead built a local folder-based system (myPKM) to maintain persistent context with AI agents. He shares an audit revealing systemic issues like contradictory documentation, forgotten rules, and parallel session conflicts, then outlines how to systematically fix these problems using the ICOR productivity methodology.
I do not download invoices anymore. My folder and Claude do it.
The speaker demonstrates how he uses AI (Claude) integrated with his local folder system to automate invoice collection for his accountant, downloading 36 invoices in 5 minutes with minimal manual intervention. He explains his AI team methodology, work streams, and how persistent local knowledge bases enable AI to handle recurring tasks independently without constant re-explanation.
Claude finished the work. Then I lost it in the chat.
A creator demonstrates a comprehensive local-folder-based system for managing AI-generated deliverables asynchronously, featuring an interactive browser interface that tracks decisions, changes, and conversations without relying on external databases or web applications. The system enables collaborative editing, visual annotations, and automatic context preservation across multiple parallel AI projects.