Subscribe
Sign in
Home
Notes
Archive
About
Your AI Should Earn the Right to Interrupt You
Always-on assistants can keep checking. The useful ones know which changes deserve your attention.
Oct 5
•
Payton Bilodeau
1
September 2026
Why AI Can Reuse Work and Still Forget
OpenAI’s new caching update makes repeated context cheaper to process. That helps explain why faster AI and better memory are separate promises.
Sep 28
•
Payton Bilodeau
1
How to Think Clearly in the Era of AI:
A threaded essay of maxims on thinking clearly explained.
Sep 21
•
Payton Bilodeau
1
Asking ChatGPT to Think Harder Isn’t Enough
A change to ChatGPT’s defaults makes the thinking setting worth checking. Here’s when to raise it, and when the problem is somewhere else.
Sep 21
•
Payton Bilodeau
1
Stop Squeezing Every AI Request Into a Text Box
This week’s image, voice, and file updates make it easier to show what you mean, provided you choose the right input for the job.
Sep 14
•
Payton Bilodeau
2
1
Give Your AI a Job Worth Finishing
Astra and Fable 5.1 target longer projects. The useful test is whether they can leave you with something you can actually use.
Sep 7
•
Payton Bilodeau
4
How to Think Clearly in the Era of AI
Easy answers can make us feel smarter without making us wiser. Three ways to keep thinking for yourself.
Sep 5
•
Payton Bilodeau
1
1
Keep a human on the click
They shipped memory, a Slack trigger, and a real browser
Sep 1
•
Payton Bilodeau
4
1
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts
