No, because Claude always agrees with Claude. If you give it Claude input you have to massively fight it to admit it’s wrong.

Ask me how I know.

That might be a viable defense in one scenario, but as a global approach in the context of everything currently happening, it feels more like a "tools of the enemy" issue. Like combating misinformation with more misinformation, it will backfire sooner or later. And it profits the broligrarchy the whole time you're doing it.

But you should add a disclaimer that you will not be held responsible for anything that is said in this communication, because you did not write it or read it

That's exactly what the article suggested...

Another strategy is to use AI to fight AI. This is a good one for handling workslop from managers. You can do this in two broad ways. First, instead of carefully reading it, paste it into an LLM of your own and ask for a short list of the salient points. Second, you can sometimes simply ask an LLM for an entire response. In a sense, this makes you part of the problem, so I can see why some people might be uncomfortable with it. But it’s more sustainable than spending ten minutes of your effort for every ten seconds of theirs.