Mike's Memos

Mike Litman

Short sharp takes on tech, culture, AI, and building. Each episode is a five-to-eight minute essay from mikelitman.me/memos, read aloud. By Mike Litman, BIMA 100 Tech Pioneer, published author, 15+ years in digital, now building AI-native products.

  1. Apr 23

    The Second Brain Has Overtaken The Second Screen

    18 slides tracing the arc from second screen to second brain: why the first concept dissolved, what replaced it, why AI changed the contract, and who is building the mainstream version. "The second screen split your attention. The second brain extends it." Key takeaways: - 98.5% of the potential market has not adopted a true second brain system: Notion has 100 million users, Obsidian around 1.5 million - LLMs changed the contract: before, the second brain was a searchable filing cabinet; now it can reason across everything you’ve captured - The mainstream version will win by removing the setup cost entirely: the second brain disappears, like HTTPS, until it just works Full memo: https://mikelitman.me/secondbrain

    8 min

  2. Apr 22

    1,000 Venues. One Question.

    An AI called 10,000+ London venues with one question. Here is what broke, what got fixed, what the data shows, and why a dataset that didn’t exist is now a business. "This data should have existed a long time ago. I made it for myself. Now it does." Key takeaways: - A dataset that didn’t exist can be built by one person with a voice AI, a phone number, and a well-scoped question at scale - The amber category (limited access / nuanced) is nearly as large as green — that nuance is the moat. No competitor can manufacture it without making the same calls. - Self-improving data pipelines are the next step beyond static collection: 184 venues appeared overnight without a single new call, because the system corrected its own past decisions. Full memo: https://mikelitman.me/buggysmart1000

    8 min

  3. Apr 22

    Campaigns End. Institutions Compound.

    Google is not trying to sell Pixel to creative professionals. It is trying to become the infrastructure of their practice. Campaigns end. Positions compound. "A product launch is a season. An institution is a decade." Key takeaways: - Google is not renting cultural permission with PIFT -- it is trying to become load-bearing infrastructure for the next generation of creative professionals - The brand that trains designers before they decide their tools wins the next decade of wearables, not the brand with the best spec sheet - PIFT's cultural credibility is co-owned by Highsnobiety, making it structurally harder to kill than a Google-owned product Full memo: https://mikelitman.me/credibility-pift

    7 min

  4. Apr 21

    Moving Fast Breaks Things Slowly

    20 slides on why moving fast with AI creates a new category of failure: errors that don't crash, don't alert, and accumulate across months of production. The morning audit that found 64 silent issues, the case for encoding quality into CI instead of human memory, and what 'done' actually means when the build is the check. "The faster you ship, the harder the rails need to be." Key takeaways: - Speed doesn't create fewer mistakes, it creates invisible ones: wrong fields, dead URLs, broken paths that pass the build and look fine until you actually look - Checklists degrade under pressure. A checklist you sometimes forget to follow is not a quality system, it's a quality wish - The answer is to encode invariants into the machine: CI that rejects on every push, required fields that block without explanation, validators that don't rush and don't care how good the talk was - Done used to mean shipped and works. In fast AI-assisted production, done means shipped, works, and the system will catch it if it drifts Full memo: https://mikelitman.me/movingfast

    7 min

  5. Apr 17

    50 Decks Before Claude Design

    22 slides on building 50 presentations before Claude Design existed: the curation system, the nine-step production pipeline, the shared template infrastructure, the content engine that turns one deck into four formats, and the time collapse from two weeks to hours. Plus first-hand reactions from two Anthropic insiders on launch day. "Better tools lower the cost of a bad idea. Only better thinking raises the floor." Key takeaways: - The curation is the hardest part — not everything that lands in the inbox becomes a deck, and the editorial gate is what makes 50 presentations feel coherent rather than noisy - The nine-step pipeline (deck, OG, thumbnail, redirect, mobile, carousel, blog, X Article, claims audit) turns one idea into four pieces of content automatically from the same source material - The thinking has not got faster — everything around the thinking has. This deck was conceived, built, and published on the day Claude Design launched. Full memo: https://mikelitman.me/beforeclaudedesign

    7 min

  6. Apr 17

    Forbes AI 50 · 2026

    The Forbes AI 50 maps the most promising private AI companies of 2026. 14 slides decode all 50 across 6 sectors -- foundation models, creative AI, vertical tools, physical AI -- plus notable absences and geographic concentration. "American by default. European by exception." Key takeaways: - The Forbes AI 50 covers private companies only -- meaning Microsoft, Google, Meta, Amazon and xAI are all invisible, which changes how you read the whole list - Physical AI is a named sector for the first time: Physical Intelligence, Skild AI, Applied Intuition and World Labs confirm robots and autonomy have arrived as investment categories - Mistral AI is the only prominent European entrant -- the overwhelming majority of the list is US-based, concentrated in San Francisco and New York Full memo: https://mikelitman.me/forbesai50

    7 min

  7. Apr 17

    The Distribution Supercycle

    15 slides on the distribution supercycle: why production costs are going to zero, why CPMs keep climbing, and why the Distribution Engineer — part builder, part strategist — is the most valuable hire of 2026. Includes the Anthropic one-person proof, the four-levels framework, and a direct prescription for founders. "Your first hire is not a head of marketing." Key takeaways: - Production costs are going to zero while attention supply is fixed — CPMs only go up, which is why smart capital is acquiring media and audience assets at record prices - The Distribution Engineer treats distribution like an engineering problem: building agents and infrastructure, not running campaigns — one person can outperform a team of ten - Taste and distribution are the only two moats AI cannot commoditise — every other advantage is one Claude update away from being replicated by a teenager with a laptop Full memo: https://mikelitman.me/distribution-engineer

    7 min

  8. Apr 17

    The Great Unplug

    Nordic countries were the gold standard of global education. They went all-in on classroom technology. The data came back negative. They reversed course. Here is what they found -- and what it means for every organisation rolling out AI right now. "The world's best schools ran the biggest EdTech experiment in history. Then they reversed it." Key takeaways: - The countries that led the world in going digital became the first to say: we got this wrong. - Paper and screen are not the same cognitive act. Screen readers skim. Print readers re-read. - Before rolling out any new tool, define what you will measure and when you will check it. Full memo: https://mikelitman.me/unplug

    7 min
See All (59)

About

Short sharp takes on tech, culture, AI, and building. Each episode is a five-to-eight minute essay from mikelitman.me/memos, read aloud. By Mike Litman, BIMA 100 Tech Pioneer, published author, 15+ years in digital, now building AI-native products.

Information

  • Creator
    Mike Litman
  • Episodes
    59
  • Rating
    Clean
  • Copyright
    © Mike Litman
  • Show Website
    Mike's Memos