I think this single fact is a major source of enshittification in large software products, especially in the era of ML/AI. If your quality is 99%, it sounds like "you have solved your task", but in reality there is a long tail that over time affects nearly every customer.

I've seen this so many times. 99% of search results are good (so within 100 queries you'll hit at least one bad result with p≈0.63), 99% of dashboard panes load normally (so a dashboard with 20 panes is broken in nearly 1 in 5 loads), and so on. If your LLM gets 99% of tool calls right, nearly every session will contain a malformed tool call.

Probabilities are hard for humans, probably.

Alternatively getting the last piece of 1% could mean 99% of the effort. Would you consider it fruitful to chase 100%?

When measuring and reporting models to the non-saavy, I usually reframe them into odds. One failure for every 49 successes is a scary failure rate when operating at a large scale.

This is largely why I don't condone LLMs in operational pipelines. Your workflow? Fine. The company's? Hell no.