Unmaking Sense

Not only textual data but the kinds of images produced by AI are suffused with cultural biases and prejudices. We need to be aware of the distortions that can become ingrained if these implicit biases are not countered and held in check.

We have not even started to come to terms with the motivational implications of AI, whose dangers far outweigh the more dramatic science fiction scenarios so many love.

We talk about ways to upload whole files, text files as prompts to locally running large language models, and after that we talk about creating false videos that look as though you’re saying things that you’re not saying, and have never said, on the basis of little more than about 30 seconds worth of data.

We explore some of the intriguing parallels between diffusion models - noising and denoising - and human learning and recall with emphasis on innovation and creativity.

A short introduction to RAG.

Unmaking and remaking sense as a means of creating new worlds.

A series of placeholders for what is yet to come in this extraordinarily rich new vein of thoughts, thinking and latent spaces.

An unexpected light illuminates the whole Unmaking Sense series. A key-stone falls into place; understanding occurs, emerging from the fog of 450 episodes.

How diffusion models work in general outline, what that tells us about neural nets and the AI process, and what it suggests we might usefully learn from the whole process about the operations of the human brain. Some passing remarks about copyright law.

We explore one aspect of understanding that perhaps beguiles us into imagining that t is some special quality: our inside story. And we consider how AI may come to experience something similar in an occasional, emergent way.