Latestai
Solving the Memory Problem
I built a persistent memory layer that sits outside any single AI provider. Now every tool I use shares the same context — and it's changed how I work entirely.
4 min readRead Essay →
The best AI doesn't replace thinking.
It makes room for more of it.
Series
Ai Skills Series
2 parts
12
AI feels new. The labour pattern isn't.
Every few decades, a technology arrives that triggers the same prediction - this one will finally make people obsolete. The technology changes, but the mistake doesn't.
Prompting Is Not the Moat
Prompt engineering courses are booming, but the skill they teach is depreciating fast. The durable value in AI sits above the interface.
The Alignment Series
3 parts
123
The Alignment Problem Is a Human Problem
AI alignment research assumes humans can specify what they want. Behavioural science says otherwise.
AI Is Making Up Its Own Values
AI models are developing coherent internal value systems. Some of those values are ones we wouldn't choose.
DeepSeek thinks it's Claude
An AI that doesn't know who it is turned out to be a fingerprint of industrial-scale model distillation - and the ethics are more complicated than anyone wants to admit
All Writing
Full Archive →Archive
Previous
