Rand兰德：2026年保护高级人工智能的四种治理方法报告（英文版）

To draw lessons for the AI industry, the researchers examined security governance approaches in seven high-risk sectors, such as nuclear, chemical, and health care.1 Among these industries, they found that federal agencies and industry consortia established compliance regimes to promote industrywide adoption of security standards. By combining incentives and penalties, these governance models sought to shift firms’ cost-benefit calculations and drive greater investment in protective measures.