Anthralytic’s Substack
Subscribe
Sign in
Home
Podcast
Notes
Archive
About
I Broke the App I Built With AI to Teach How AI Breaks
There is a new card in the How AI Breaks in Social Impact app I built.
Jul 9
•
Anthralytic
1
1
I Used AI to Help Write This Piece. Judge the Output.
Disclosure: I used an AI assistant as a conversation partner while writing this piece.
Jun 29
•
Anthralytic
12:07
Audio Version - A Hammer Looking for a Nail: Anthropic’s Nonprofit Upskilling Program Repeats International Development's Mistake
Anthralytic is a strategy and evaluation studio for mission-driven organizations.
Jun 28
•
Anthralytic
1
2
Most Popular
View all
Aid's Reckoning Is Here. Let's Not Waste It
Jun 3, 2025
•
Anthralytic
7
1
1
The Mountain You Don't Have to Climb
Jan 12
•
Anthralytic
6
1
Data Security When Security Makes You Suspicious
Feb 9
•
Anthralytic
5
3
Effective Altruism Got Weird—But Not All of It Should Be Thrown Out
Jun 18, 2025
•
Anthralytic
3
Social Impact Data Vulnerability in Three Acts
Jul 25, 2025
•
Anthralytic
5
1
USAID, One Year Later: The System They Broke and the People Still Counting
Mar 10
•
Anthralytic
3
1
2
Recent posts
View all
A Hammer Looking for a Nail: Anthropic’s Nonprofit Upskilling Program Repeats International Development's Mistake
Anthropic is spending one hundred fifty million dollars to put a thousand early-career workers inside nonprofits.
Jun 26
•
Anthralytic
1
6:48
Audio Version: Who Should Decide Which AI is Too Dangerous?
Thanks for reading Anthralytic’s Substack!
Jun 23
•
Anthralytic
1
Who Should Decide Which AI Is Too Dangerous?
On June 9, 2026, Anthropic launched Claude Fable 5.
Jun 23
•
Anthralytic
1
8:58
Audio Version: Three Moves to Tell If Nonprofit AI Works (and for whom)
AI is being adopted across the nonprofit sector inside a frame that never asks who benefits. A question to hold, a practice to start, a demand to make.
Jun 11
•
Anthralytic
1
Three Moves to Tell If Nonprofit AI Works (and for whom)
AI is being adopted across the nonprofit sector inside a frame that never asks who benefits. A question to hold, a practice to start, a demand to make.
Jun 8
•
Anthralytic
1
Grading Their Own Homework
The nonprofit sector's evidence base on AI is being produced by the companies selling it.
Jun 3
•
Anthralytic
1
5:39
Audio Version: Measuring the Wrong Things Faster
Thanks for reading Anthralytic’s Substack!
Jun 1
•
Anthralytic
Measuring the Wrong Thing Faster
AI is accelerating nonprofit work. The dominant report mistakes the acceleration for impact.
Jun 1
•
Anthralytic
5:27
Bonus: How I Accidentally Created an Agentic Text-to-Audio Workflow
This was supposed to be a simple question.
May 31
•
Anthralytic
1
See all
Anthralytic’s Substack
Anthralytic Substack
Subscribe
Recommendations
View all 14
Thinking Freely with Nita Farahany
Nita Farahany
Business English for Social Impact (BE4SI)
Ryan Turner
An Africanist Perspective
Ken Opalo
AI Governance, Ethics and Leadership
AI Governance Lead ⚡
The VoxDev Substack
VoxDev
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts
