I'm also confused about the anonymous interview scenario. Does this actually occur enough to require a complex rule?
If we're using text, why use accents instead of just labelling the text as "Participant A", or similar, like we already do for subtitles when disambiguation is required?
If we're using voices morphed to sound the same, could we similarly have a visual indicator of who is talking? If no visuals are possible, use different morphs so they sound different. If that's not possible, just conduct the interview by addressing participants with their pseudonym so we know which response to expect. Like anyone would naturally do.
This feels like a solution looking for a problem.
You also noticed he's suddenly talking about securing rare earths every day now? I'm hoping someone asks him why they're so important because he clearly doesn't know what they are, and is just following orders from Twitler.