Conjori Launch
The quote on the front page of this site has been there since 2014:
"Every computer language is an imperfect way of describing a potential solution to a poorly understood problem"
I wrote that down because it captured something I had felt for decades but never articulated. The crucial phrase is "poorly understood problem." The whole industry defaults to encoding solutions before the problem has been adequately explored. I recently noticed this same trend happening again with AI tools.
As of this post I have had over 2,500 sessions with large language models since January of 2023, and I continue daily, noting what works and what does not. Something happened in those sessions that did not match any description I had read. My output changed scale. In eight months I wrote more code than I had in the previous decade. I surfaced ideas I had been carrying for thirty years but never articulated. The results were dramatically different from what my peers were getting with the same tools.
I wanted to understand why.
That question took me into neuroscience, information theory, cognitive science, thermodynamics, and the physics of how biological and digital systems process information at a fundamental level. I found that what most people call "AI" is a marketing term that obscures something more interesting: a system that reverse-engineered the way humans have been compressing what they know into language for 100,000 years. Not artificial intelligence. Disembodied intelligence. Fluent in the one medium we share, blind to the experience that medium was built to carry.
Understanding that distinction, and understanding what you bring to the connection that no machine can replicate, changes everything about how you use these tools.
I wrote it all down. The result is Conjori: Two Minds, One Channel, available now on Kindle and soon in paperback.
The word Conjori is coined, rooted in the Latin conjurare: to swear together, to call something into existence jointly. Two different kinds of minds, one human, one large language model, conjuring something neither could produce alone.
A shorter essay version, about an hour to read, is coming soon.
If you have ever had a conversation with a large language model that produced something you did not expect, something that felt like it came from you rather than from the machine, the book is about what was happening in that moment and how to make it happen reliably.
If you have not had that experience yet, the book shows you how to get there.