
Nano Thoughts
The Church of Recursive Self-Improvement
On the surface, Silicon Valley’s AI boom looks like a classic venture capital money fever.

Nano Thoughts
On the surface, Silicon Valley’s AI boom looks like a classic venture capital money fever.


Nano Thoughts
In a recent X post, Chamath warned consulting firms that deploying Anthropic or OpenAI directly into their organizations is “letting the fox into the hen house.” That warning feels sharper after OpenAI launched the OpenAI Deployment Company, a majority-owned venture built to help organizations deploy AI systems, backed by $4 billion of initial investment and partners including TPG, Advent, Bain Capital, Brookfield, Bain & Company, Capgemini, and McKinsey (




Nano Thoughts
The strange thing about better AI is that it may not reduce the need for human judgment.



Nano Thoughts
The next chapter of enterprise AI will not be one assistant that knows everything.


Nano Thoughts
We did not start by calling it Company Brain.






Nano Thoughts
Companies do not only forget what happened or why a decision was made.


Nano Thoughts
Almost everything important in a company happens in meetings, messages, or emails.


Nano Thoughts
In the first piece, I argued that a real Company Brain needs three kinds of memory: factual memory, interaction memory, and action memory.


Nano Thoughts
One of the hardest parts of any organization is institutional friction.


Nano Thoughts
TL;DR The augmenting and replacement futures use the same models and the same tools.

Nano Thoughts
TL;DR Only 4 out of 128 dimensions in the KV cache carry meaningful signal.


Nano Thoughts
In the last few weeks, the AI/Tech community seems to have collectively realized what biology knew all along, that memory is a graph, not a database.

Nano Thoughts
The winners of the next decade won’t just build better AI tools.

Nano Thoughts
One of my earliest memories from childhood is that of my grandmother smearing a thin slurry of fresh cow dung, from our farm, and water and swept it over the red earth.

Nano Thoughts
Biostate AI's goal is to get 90% of the world's people to live to 90 years old. This requires global collaboration to train the AI.

Nano Thoughts
Limitations are strengths.


Nano Thoughts
Top-down approaches to medicine are needed to complement bottom-up cell/organoid methods