This market resolves to Yes if a credible public demonstration or test confirms the ability of AI to self-replicate in a controlled environment using a vulnerability test by July 1, 2026. The test must be acknowledged by at least two reputable cybersecurity or AI research organizations.