Trump signs executive order to review AI models before they’re released
Trump signs executive order to review AI models before they’re released
特朗普签署行政令，要求在 AI 模型发布前进行审查
President Donald Trump signed an executive order Tuesday creating a “voluntary framework” for AI companies to share their frontier models with the federal government before they’re released “to promote secure innovation and strengthen the cybersecurity of critical infrastructure.” 唐纳德·特朗普总统周二签署了一项行政令，建立了一个“自愿框架”，旨在让 AI 公司在发布前与联邦政府分享其前沿模型，以“促进安全创新并加强关键基础设施的网络安全”。
The order says the US AI industry has succeeded in part “because we refuse to stifle this innovation with overly burdensome regulation,” but that it also recognizes new AI capabilities come with security risks. Accordingly, it directs several federal agencies to come up with a framework to “assess the advanced cyber capabilities of AI models” before they’re released to the public. Companies would have the discretion of whether to share their models with the government pre-release, but could get certain confidentiality protections if they choose to do so. It also requires the federal government to prepare cyber defenses for AI, especially for critical infrastructure. 该行政令指出，美国 AI 行业的成功部分归功于“我们拒绝用过于繁琐的监管来扼杀创新”，但也承认新的 AI 能力确实带来了安全风险。因此，该命令指示多个联邦机构制定一个框架，在 AI 模型向公众发布前“评估其先进的网络能力”。公司可以自行决定是否在发布前与政府分享模型，如果选择分享，则可获得一定的保密保护。该命令还要求联邦政府为 AI 准备网络防御措施，特别是针对关键基础设施的防御。
The order comes after Trump postponed at the last minute signing a previously planned executive order that he worried could “get in the way” of competing with China. While the earlier version would have allowed AI companies to voluntarily share their models 14 to 90 days before release, according to The New York Times, the current version asks companies to share their models up to 30 days before public release. Google, Microsoft, and xAI agreed last month to allow pre-release review by the Commerce Department’s Center for AI Standards and Innovation (CAISI). 在此之前，特朗普曾于最后一刻推迟签署一项原定的行政令，因为他担心该命令可能会“阻碍”与中国的竞争。据《纽约时报》报道，早先的版本允许 AI 公司在发布前 14 到 90 天内自愿分享模型，而当前版本则要求公司在公开发布前最多 30 天内分享模型。谷歌、微软和 xAI 上个月已同意接受商务部下属的 AI 标准与创新中心（CAISI）的发布前审查。
OpenAI and Anthropic had already agreed to share their models with CAISI back in 2024 under President Joe Biden as part of Biden’s push for AI safety guardrails. But until recently, the Trump administration downplayed safety concerns and took a hands-off approach under former White House AI czar David Sacks. The order signed Tuesday explicitly says it shouldn’t be taken as a form of mandatory licensing or preclearance. Still, it reflects some willingness of the Trump administration to employ oversight of AI companies. OpenAI 和 Anthropic 早在 2024 年乔·拜登总统任期内，就作为拜登推动 AI 安全护栏计划的一部分，同意与 CAISI 分享其模型。但直到最近，在白宫前 AI 主管大卫·萨克斯（David Sacks）的领导下，特朗普政府一直淡化安全担忧，并采取了不干预的态度。周二签署的行政令明确表示，这不应被视为强制许可或预先批准的一种形式。尽管如此，这反映出特朗普政府在对 AI 公司实施监管方面表现出了一定的意愿。
“The White House is officially Mythos-pilled” “白宫正式被 Mythos‘洗脑’了”
One factor in this shift may be Anthropic’s limited April rollout of its powerful Mythos model, which the company said had flagged “thousands of high-severity vulnerabilities, including some in every major operating system and web browser.” Mythos also seemed to create an opening to thaw tensions between Anthropic and the administration, following its legal battle with the Pentagon over its use of AI for autonomous lethal weapons and mass surveillance. 这一转变的一个因素可能是 Anthropic 在四月份对其强大的 Mythos 模型进行的有限度发布。该公司称，该模型标记了“数千个高严重性漏洞，包括每个主流操作系统和网络浏览器中的漏洞”。在 Anthropic 因五角大楼使用其 AI 技术开发自主致命武器和进行大规模监控而引发法律纠纷后，Mythos 的发布似乎也为缓解 Anthropic 与政府之间的紧张关系创造了契机。
The newly signed order has so far garnered praise even from groups that have advocated against restrictions on state AI laws. “The White House is officially Mythos-pilled,” Americans for Responsible Innovation President Brad Carson said in a statement, adding that the order shows the Trump administration is taking AI vulnerabilities seriously. Alliance for Secure AI CEO Brendan Steinhauser said his group is “pleased to see that the Trump administration is taking the risks of these models seriously,” and both Steinhauser and Carson urged Congress to codify mandatory protections. 这项新签署的行政令目前已获得赞誉，甚至包括那些一直反对限制州级 AI 法律的团体。“白宫正式被 Mythos‘洗脑’了，”美国负责任创新组织（Americans for Responsible Innovation）主席布拉德·卡森（Brad Carson）在一份声明中表示，并补充说该命令表明特朗普政府正在认真对待 AI 漏洞。安全 AI 联盟（Alliance for Secure AI）首席执行官布伦丹·斯坦豪瑟（Brendan Steinhauser）表示，他的团队“很高兴看到特朗普政府认真对待这些模型的风险”，斯坦豪瑟和卡森都敦促国会将强制性保护措施编入法律。