One thing surprised me after learning how @GenLayer reaches a verdict. It doesn't expect independent AI models to sound the same. It expects them to mean the same thing. That's a subtle difference, but it changes everything. If five experts reach the same conclusion using different words, I don't trust them less. I trust the conclusion more, because nobody copied anyone else's reasoning. That's exactly why checking meaning instead of wording makes so much sense to me. Consensus shouldn't reward imitation. It should reward independent agreement.