OpenAI 近期就前沿人工智慧开发的放缓协调是否具备合法性向美国国会寻求指引。OpenAI 首席科学家 Jakub Pachocki 主张，业界协调放缓研发步伐对于保障自我改进型 AI 系统的安全至关重要，但法律学者指出，企业间联合暂停开发可能被视为限制产出，进而面临违反《谢尔曼反托拉斯法》等反垄断法规的风险。

为了解决此法律疑虑，美国两院跨党派议员于今年七月提出了《对抗威胁与安全风险合作法案》，旨在明确允许 AI 实验室在安全与防护领域开展协调合作而不受反垄断法处罚。然而，该法案目前仍处于司法委员会阶段，且在即将到来的期中选举前可能难以迅速完成立法。

反垄断担忧并非阻碍合作的唯一原因，商业市场竞争、中美科技竞争以及对安全发展路径的分歧皆让各大实验室难以携手。前 OpenAI 共同创办人 John Schulman 等业内人士批评反垄断问题仅是借口，而近期一连串安全事件与研究员对人类面临生存风险的公开警告，正进一步加剧国会对 AI 监管的迫切需求。

OpenAI has recently sought guidance from members of the US Congress regarding whether coordinating an industry-wide slowdown in frontier AI development would be legally permissible. OpenAI’s chief scientist, Jakub Pachocki, contends that coordinating to pace future progress is essential for safeguarding self-improving systems, yet legal scholars caution that a collective development pause could be construed as restricting output under the Sherman Antitrust Act.

In response to these concerns, a bipartisan group of lawmakers introduced the Collaboration on Adversarial Threats and Security Risks Act in July to explicitly exempt AI safety and security coordination from antitrust statutes. Although advocacy groups note a growing congressional interest in addressing AI risks, the legislation remains pending in the Judiciary Committee and is unlikely to pass before the upcoming midterm elections.

Beyond antitrust liabilities, intense market competition, national security rivalry with China, and differing safety philosophies also prevent major labs from collaborating effectively. Industry insiders like John Schulman argue that legal risks serve merely as convenient excuses for rivalry, while recent security incidents and whistleblowing warnings from researchers continue to amplify urgent legislative calls for AI oversight.