korrents
Pieter Levels

What Pieter Levels thinks about LLMs

@pieter-levels · 69 positions · 2 changes of mind

Dutch bootstrapper who builds and runs his internet startups alone — Nomads.com, Remote OK, Hoodmaps, Photo AI and Interior AI — and has written about it at levels.io since 2013. Posts as @levelsio.

Everything they publish, on ppll ↗

Pieter Levels did not write this page.

We collected these quotes from things they published elsewhere, and every quote links to where it was said. They have no account here and have not endorsed this site. Quotes are word for word; the short line under each one is our own restatement, not their wording. Their own site. Is this you? Claim it or ask us to remove it. Or tell us what is wrong here.

4 dated positions, 2024 to 2026, in their own words. Our reading of what Pieter Levels has said — not written or endorsed by them.

4 positions so far — this page is not yet offered to search engines.

  1. Our readingEvery email should be routed through an LLM that scores it and approves or rejects it, with what is left ordered by importance.

    Their wordsThere is no reason NOT to have every single email you get routed through an agent / LLM / API to be scored and approved or rejected And then the approved emails ordered by importance It's crazy this isn't a feature of Gmail, Fastmail etc AI could permanently fix email!

    ↗@levelsio on Xx.com

    Interesting

  3. Our readingEvery large language model is trained on billions of pages of copyrighted data

    Their wordsAlso the irony knowing that every single LLM is trained on billions of pages of copyrighted data

    ↗I keep getting rejected by Claude for the most benign things nowlevels.io

    Interesting
  4. 3 weeks earlier

  5. Our readingIf nobody creates content any more because everyone just asks an AI, there is nothing left for AI to train on.

    Their wordsThe problem is if nobody creates new content anymore and just asks their AI every question, the AI has nothing to train on anymore

    ↗The Dead Internet Theory could become real if AI replaces all human content creationlevels.io

    Interesting
  6. 24 months earlier

  7. Our readingAn LLM moderates an online community better than human moderators do, because it understands humour and double meanings.

    Their wordsI have GPT-4o and it’s really, really, really, really good. It understands humor. You could say something bad, but it’s like a joke and it’s not offensive so much so it shouldn’t be deleted. It understands that.

    ↗Pieter Levels: Programming, Viral AI Startups, and Digital Nomad Life | Lex Fridman Podcast #440youtube.com 19th of 30 in this recording

    Interesting