My website: boas.work
Boas
Karma: 1
Boas’s Shortform
Following the spirit of digital philosophy (like the kind of Gregory Chaitin), would defining rationality as “compression algorithm of language using human language” be accurate?
In the sense that prediction can be seen as compression in Algorithmic Information Theory, so writing down/”predicting” conclusions from previous texts is the same as compressing those texts
Very intriguing story! It leaves some questions and open ends, like:
- What happend to mnt? What is the fate of the ship? The “charge fault” number keeps incrementing (30, 32, 35, 41...).
- How did it figure out that it was a ramscoop so quickly? My friend suggested an explanation: the model weights are designed/trained for the galactic context
- Where does the symbol ⚶ come from? Why does it originate from DTENSOR_REPLACE_ZERO? Is ⚶ at the origin of the embedding space?
A friend shared me this story after I shared this CoT from Gemini 3.1 Pro used inside GitHub Copilot:It is amazingly similar
If we are talking about human to human interaction, I think we cannot speak of a binary divide between monitoring and not-monitoring. In some sense, humans are always passively monitoring. For example, if parents hear screaming from their children, they will often investigate/act even though they weren’t actively monitoring. More generally, monitoring and steering happens in all interactions slightly