I’ve been somewhat critical of the direction of OpenAI recently. After its ‘code red’ scenario and a sweep of seemingly bad decisions, it seemed to me to be losing its footing. But its latest venture ...