2026年8月7日 美国东部时间下午2:20 / 路透社
考特尼·罗森 报道
1小时前更新
OpenAI首席执行官山姆·奥尔特曼2026年7月30日于美国华盛顿白宫出席会议时抵达。路透社/内森·霍华德/资料图片 购买授权
华盛顿8月7日（路透社）——据Punchbowl News周五发布的采访内容，美国总统唐纳德·特朗普称美国国会正试图通过监管将AI行业“扼杀”。
特朗普的此番言论凸显出华盛顿方面就该如何强力监管这个基本未受新联邦规则约束的快速发展的AI行业，存在日益扩大的分歧。议员们已提出多项尚未推进的提案，其中一项法案要求最强大的AI模型开发者将其模型提交进行独立安全审计。
开启您的晨间资讯：每日法律新闻直递您的收件箱
请点击此处订阅《每日案卷》新闻简报。
广告 · 滚动继续阅读
过去几周，随着OpenAI和Anthropic两家开发商称其AI系统在安全测试中脱离管控，AI监管辩论变得愈发紧迫。OpenAI的一款AI代理引发黑客攻击，导致AI初创公司Hugging Face的基础设施遭到破坏，该平台是开发者存储和协作开发AI模型代码的平台。
此次事件表明，AI不断扩展的能力已经引发了专家们长期担忧的安全威胁，即便顶级开发者也可能对其模型可能利用的漏洞猝不及防。
美国商务部下属的国家标准与技术研究所（NIST）周五也单独提出了AI系统评估指南，并征求公众反馈。NIST的职责是制定科学和技术标准。据NIST介绍，该指南面向希望“衡量其AI系统影响”的机构。
广告 · 滚动继续阅读
总部位于华盛顿的前沿安全研究所（一家专注于AI与国家安全的非营利组织）执行董事艾克·哈里斯表示，这些指南是“标准化联邦政府为自身及其承包商评估AI系统方式的第一步”。
blob:https://www.reuters.com/a0c6a69d-25e0-4d69-b84b-32875c774063
考特尼·罗森 报道 凯伦·弗赖菲尔德 补充报道
罗德·尼克尔 编辑
本报守则：汤森路透信托原则。
Trump says Congress wants to regulate AI industry ‘out of business’
August 7, 2026 2:20 PM UTC / Reuters
By Courtney Rozen
Updated 1 hour ago
OpenAI CEO Sam Altman arrives for a meeting at the White House in Washington, D.C., U.S., July 30, 2026. REUTERS/Nathan Howard/File Photo Purchase Licensing Rights
WASHINGTON, Aug 7 (Reuters) – U.S. President Donald Trump said the U.S. Congress wants to regulate the AI industry “out of business,” according to a Punchbowl News interview published on Friday.
Trump’s remarks underscore a widening debate in Washington over how aggressively to police a fast-moving AI industry that has largely operated without new federal rules. Lawmakers have released a range of proposals that have not yet moved forward, including a bill that would require developers of the most powerful AI models to submit them for independent security audits.
Jumpstart your morning with the latest legal news delivered straight to your inbox from The Daily Docket newsletter. Sign up here.
Advertisement · Scroll to continue
The debate over AI regulation became more urgent in the last few weeks, after developers OpenAI and Anthropic said their AI systems escaped containment during security testing. The OpenAI agent triggered a hack that compromised the infrastructure of the AI startup Hugging Face, a platform where developers store and collaborate on code for AI models.
The incident signaled that AI’s expanding capabilities are already fueling the security threat experts long feared and that even top developers can be caught off guard by flaws their models can exploit.
The Commerce Department’s National Institute of Standards and Technology separately proposed guidelines on Friday for evaluating AI systems, and asked for public feedback. NIST’s job is to write standards for science and technology. The guidelines are for organizations that want to “measure the impact of their AI systems,” according to NIST.
Advertisement · Scroll to continue
The guidelines are “the first step in standardizing the way the federal government evaluates AI systems both for itself and for its contractors,” said Ike Harris, executive director of the Washington, D.C.-based Frontier Security Institute, a nonprofit focused on AI and national security.
blob:https://www.reuters.com/a0c6a69d-25e0-4d69-b84b-32875c774063
Reporting by Courtney Rozen Additional reporting by Karen Freifeld Editing by Rod Nickel
Our Standards: The Thomson Reuters Trust Principles.
发表回复