On Interpolatable Archives

I’ve been throwing around the term “Interpolatable Archives” for quite some time when talking about Language Models and Artificial Intelligence, and I finally got around to write down what I mean by that. This essay is a summary how i feel about AI at the moment: They are not bullshit-machines, the "stochastic parrot"-metaphor is out of date, they produce meaningful synthetic text from linguistic practice, they automatically close knowledge gaps through interpolation and are cognitive catalysts, which is precisely why they are powerful and dangerous at the same time.

It's a huge text of more than 14k words, and on Substack, I split it up into four parts for convenience. I also collected those essays into one giant thing, and for those of you who hate Substack, i published the text on Wordpress aswell.

Here are the individual parts as they were published, including AI-generated illustrations:

Part 1: The Skeleton Library (S'ouvre dans une nouvelle fenêtre) - Compulsions to Connect, Warburg, Borges and Goldsmith, Cultural Technologies and Digital Oralities

Part 2: Explosion Drawings (S'ouvre dans une nouvelle fenêtre) - Science Sans Discoveries, Textrotating Cognitive Catalysts and Exploding Your Intelligence by the Method of Warburg

Part 3: Pitfalls of Probability (S'ouvre dans une nouvelle fenêtre) - Accelerations, Anachronisms, Wishfulfillments, Severances and Homogenizations

Part 4: The House of Polly (S'ouvre dans une nouvelle fenêtre) - Useless Bullshit, Meaning Of The Poetic Kind, Sloptimizations, Thinking In Vacuums, Remainder Criticisms and Games of Chess

Here's the essay in one chunk (S'ouvre dans une nouvelle fenêtre), and here's the same text without AI-generated illus (S'ouvre dans une nouvelle fenêtre), and here's the essay on Wordpress (S'ouvre dans une nouvelle fenêtre).

I worked on this piece quite some weeks after spending the past year reading a lot, taking notes and thinking about the concept of AI as interpolatable archives. I'll sure use this concept in the future to branch off into various topics i hit on in the essay, the first one being about why AI aka interpolatable archives are Intelligence Minus Cognition (S'ouvre dans une nouvelle fenêtre).