Musk Says Many Will Be ‘One-Shotted’ by Super Persuasive Superintelligence

The remark is less a technical claim than a signal about how AI risk is being framed by one of the industry's most-followed voices: not runaway capability or physical harm, but persuasion as the attack surface. That framing matters because it shifts attention toward model behavior and human susceptibility rather than compute thresholds or safety benchmarks, a lane regulators and platform operators have barely begun to address. The post's vagueness is itself notable — no systems, mechanisms or timeframe are named, leaving the claim unfalsifiable. Whether this persuasion-focused risk language gains traction among policymakers and labs is the open question.