# @voooooogel — 2024-03-12

♥186 ↻9 · https://x.com/voooooogel/status/1767546221051408711

if you ask an llm to summarize, remember that while the result may be a condensed form of the source text, it isn't really a summary.

the llm is just a sophisticated next token predictor *imitating* summarization—the result text is only a very convincing simulacrum of a summary.

tags: author:voooooogel, kind:tweet, on:observations, year:2024
cited on: observations