Models will only get better

About: Every Saturday, I share my thoughts on working with AI. Subscribe for free.

How do you get better at what you do?

By focusing on the details. Each time you make one thing slightly better, the overall result gets better. You never notice it until you do. Time passes, and now you no longer recognize where you started.

LLM model improvements seem to follow this rule. You use them every day, notice nothing, and one day you get an answer you really didn't expect. That's it, the model got better overnight.

But it's not magic, labs are working tirelessly on details to make the models better. And because of this, the models you're using today are the worst you'll ever use.

Models will only get better.

Enjoy.

Recent posts
featuredI made a CLI to enrich French-English phrasebooks with AI translations, audio, and images
See how I used the OpenAI API to generate audio and images
code
Think for yourself before you ask
Make the effort and your mind will reward you
Improve your docs by giving your AI assistant the project's issues
See why a virtual keyboard bug calls for issue-aware docs
misc
Curious about the tools I use?
Eden
My ChatGPT client for the Emacs text editor.
One.el
My minimal static site generator.
Free course
Build a ChatGPT client in Emacs Lisp.