Reframing Superintelligence: Comprehensive AI Services as General Intelligence — LessWrong