OpenAI's Sam Altman and the father of quantum computing just agreed on a Turing Test 2.0 (www.businessinsider.com)

At a Berlin fireside chat, OpenAI CEO Sam Altman and physicist David Deutsch proposed a concrete new benchmark for artificial general intelligence: if an AI can solve quantum gravity and plausibly explain the story of how and why it pursued that solution — the problems it chose, the reasoning and tests it used — that would count as human-level intelligence. Deutsch, long skeptical that brute-force training can produce genuine minds, argued that current large language models excel at mimicry and assembling knowledge but do not “create” knowledge in the sense of identifying problems, inventing theories, testing them and improving them. Altman welcomed the hypotheticals; both agreed that a model which independently produced and justified a credible theory of quantum gravity would meet a meaningful Turing Test 2.0. The proposal matters because it shifts AGI evaluation from linguistic performance to demonstrable scientific creativity and verifiable epistemic output. Quantum gravity is an intentionally high bar: it demands original mathematics, new conceptual frameworks and empirical consequences — not just fluent exposition of existing literature. Practically, the test implies new evaluation priorities (ability to generate novel hypotheses, reproducible derivations, experimental proposals and peer-reviewable results) and raises safety and verification challenges: claims must be independently checkable, reproducible and falsifiable. Framing AGI around knowledge creation reframes research goals toward systems that can autonomously reason, design experiments and produce verifiable scientific advances.