IA 4 AI // Thought Experiments / 01

Earlier this month I asked ChapGPT:

“Could information architectures [+] be used as localised ‘programming’ for LLM’s to improve their outputs? (e.g. to make your responses more consistent, layering and building up of conversations, better synthesis etc.)”

[Some context: I’ve been trying to teach ChatGPT my Information Architecture Design framework for about the last 9 months. The aim is threefold: 1. Come to better understand these technologies by playing around with them, 2. By teaching it my IA theory, be forced to firm up the theory itself, and 3. Explore / experiment with potential applications of the two working together]

It replied in a somewhat measured tone “Yes – in theory”, and then in its most upbeat voice, throwing all caution to the wind, it continued, “Let me break down how”

I honestly can’t say if the response is profoundly insightful or total garbage, but either way, as an interactive thought experiment, I think there’s something to take out of it for anyone in our field.

Have a look:

www.jh-01.com/Downloads/IA_4_AI_Thought_Experiments_01.pdf

(Some definitions, descriptions and further reading about my IA Design work are provided at the end of the document)