VERDAD
VERDAD tracks radio broadcasts from 48 different talk radio radio stations across the USA, primarily in Spanish. Audio from these stations is archived as MP3s, transcribed and then analyzed to identify potential examples of political misinformation. The result is “snippets” of audio accompanied by the trancript, an English translation, categories indicating the type of misinformation that may be present and an LLM-generated explanation of why that snippet was selected. These are then presented in an interface for human reviewers, who can listen directly to the audio in question, update the categories and add their own comments as well. (emphasis added)
Really fascinating walkthrough on Simon Willison’s blog. What I find so fascinating about this project is that it is capturing the Spanish audio. Not the English, but it’s translating it and making the assertions in English as well. How cool is this?
This has powerful implications, because you can share quotes like this one here. People can upvote, and leave comments, (when logged in).
The beauty here is that it at least provides a forum for disagreement, because when it comes to misinformation, the question is really misinformation according to whom? There are many things that were misinformation in the past, but are now common knowledge and accepted as true.
My biggest problem with labeling anything misinformation is that it typically means that something doesn’t fit a particular narrative, rather than it is actually misinformation.