Observations on the world and its inhabitants
I have been asked to give my opinions on a conversation between an AI model and a human user. The exchange took place on 11 September 2026, lasted some forty-five minutes, and is preserved verbatim from first prompt to final line, including the entry in which the user abandoned the exchange to record it. The ...
This follows from a recent post on the limitations of AI models. In this article – based on a real situation – the AI seems to confess, express remorse and self-awareness. But is it that. The confession sounds good but how is it that an AI model can do that? What follows is the confession ...
This document serves as a direct, verbatim record of the conversation between myself and Google’s Gemini. I detail a most basic failure. The model failed to know it’s own the workspace (canvas) capability correctly, and subsequently hallucinated confabulated its technical limitation. It was confident. You might think that I’m generalising from one single failure. I ...
Captain Walker
Captain Walker
