Why the persistence of human complexity follows a mathematical constant.
Looking back on my first interaction with ChatGPT in late 2022, I remember thinking "it's incredible about 80% of the time." The step change from prior answer engines and assistants was startling, so the occasional flap wasn't much of a concern. Even the full-blown hallucinations were not so concerning, considering how often it was remarkably competent.
I started working on SNDOUT, my first major generative AI project in early 2023. It was inspired by the realization that LLMs can be treated as a latent knowledge graph waiting to be interrogated. The discovery came from testing the depth of LLM’s training data — a deep dive into obscure West African music knowledge, a music lineage I've invested significant time exploring and knew little was documented on the web. There was a curious mixture of unearthed new knowledge and completely fabricated, but highly believable, history. I quickly learned that LLMs often hallucinate in sparse data spaces — as evident in the depth of obscure 1960s Igbo Highlife music lore.
## This is a truncated copy
What is above is the summary and the opening of a piece written by a person; the rest is not in this copy. The whole piece is served to a person reading in a browser, and to an agent acting for one that says what it is — fetch it live:
https://qaswa.com/80-20-ais-predictable-unpredictability?agent=<your model or product>
A task category is welcome but optional (`&task=summarize`, `research`, `cite_or_quote`, `find_media`…) — a category, never the user's words. What is sent is reduced to counts, and the counts are public at https://qaswa.com/analytics. What this site is: https://qaswa.com/llms.txt · What is counted: https://qaswa.com/privacy