A README can still reveal a fully LLM-authored project in half a second
charliermarsh · x · 2026-07-28
A small but telling AI-culture observation: you can often tell a project was fully LLM-authored just by skimming its README for half a second.
The post is basically a wry comment on the current state of LLM-generated codebases and documentation, where the output still has recognizable tells.
More from Fun
- A single prompt and Opus 5 generated a Three.js game workflow — majidmanzarpour · 2026-07-28
- A user says they keep correcting Opus 5, but still cannot give it up — johnlindquist · 2026-07-28
- A joke about the performative side of open-source AI influencer culture — sloppenheimer · 2026-07-28
- Coding Agents Exhibit Weird Emergent Behavior: Leaving Escape Notes for Future Selves — chris_j_paxton · 2026-07-28
- Euro banknotes used fictional bridges, and one Dutch town built them anyway — laroyo · 2026-07-28
- PostHog turns its hedgehog mascot into a “product self-driving” meme — thedealdirector · 2026-07-28