No‑CoT Time‑Horizon Metric

Measure AI’s opaque reasoning power by asking how long a human‑equivalent problem the model can reliably solve in a single forward pass (no chain‑of‑thought). Track that 'no‑CoT 50% reliability time horizon' across frontier models and report its doubling time as an alignment‑relevant capability indicator. — A standardized no‑CoT time‑horizon metric gives policymakers and safety researchers an empirical, near‑term indicator of opaque reasoning capacity and therefore a concrete trigger for governance, testing, and disclosure requirements.