Welcome! 👋

One approach is to stop trying to keep up. Instead, stop and select one of his published works as a sample. Then, fact-check *every* *single* *claim* found in it, like a newspaper fact-checker would before an article is published.

Thoroughly fact-checking a sample allows you to establish a rate of falsehoods. Establishing the rate from the sample can give you some license to detach from the continuing stream: "I didn't see his new article, but I know his last one was 40% lies."

This tool helps you organize your fact-checking of a single article or transcript. It provides a visualization and statistics about the rate of falsehoods in the work.

It is not an AI tool. It doesn't assist with your research; it just helps you gather and present it.

They do! But I find those summaries less persuasive, because there's no consistancy to what they choose to fact-check. They say "we checked eleven statements in the last few years, and nine were false." This isn't quite the same as sampling a single work and determining the rate at which a person lies.

This paper, which was inspired by this article, shows that providing summaries and trends about how often a public figure lies is an effective way to change minds.