Trump’s ‘morally binding’ AI pact is not enough

On Sept.

九月。

29, President Donald Trump hosted a White House event heralding a “ new golden age .” The event followed his executive order requiring federal agencies to abandon the term “artificial intelligence” in favor of “super intelligence.” That order has more binding effect than the other news of the day—when the president and executives of half a dozen AI firms signed the White House Accord on Super Intelligence .

29 日，唐纳德·特朗普总统主持了一场白宫活动，预示着“新黄金时代”的到来。此次活动是在他的行政命令要求联邦机构放弃“人工智能”一词而支持“超级智能”之后发生的。该命令比当天的其他新闻（当时六家人工智能公司的总裁和高管签署了《白宫超级智能协议》）更具约束力。

“It’s almost like a constitution in a way,” Trump explained.

“在某种程度上，这几乎就像一部宪法，”特朗普解释道。

“The biggest people in the world signed that, and I signed it as president.” The “biggest people in the world” were Sundar Pichai of Google, Dario Amodei of Anthropic, Mark Zuckerberg of Meta, Greg Brockman of OpenAI, Elon Musk of xAI, and Jensen Huang of Nvidia.

“世界上最伟大的人物签署了它，我作为总统签署了它。” “世界上最伟大的人物”是谷歌的桑达尔·皮查伊、Anthropic 的达里奥·阿莫代、Meta 的马克·扎克伯格、OpenAI 的格雷格·布罗克曼、xAI 的埃隆·马斯克和英伟达的黄仁勋。

The president described the pact as “a form of protection” against potential threats from AI.

总统将该协议描述为针对人工智能潜在威胁的“一种保护形式”。

This was because the companies have agreed to what he called “tremendous self-policing.” This means that while the super intelligence executive order is enforceable across the government, the AI accord is only, in the words of the president , “morally binding.”

这是因为这些公司已经同意他所说的“巨大的自我监管”。这意味着，虽然超级智能行政命令可以在整个政府范围内强制执行，但用总统的话说，人工智能协议仅具有“道德约束力”。

No doubt the six industry leaders who signed were sincere in their concerns about the threats inherent in what they are building.

毫无疑问，签署的六位行业领导者真诚地担心他们正在构建的产品所固有的威胁。

Sincerity, however, does not eliminate self-interest.

然而，真诚并不能消除私利。

Self-determined and self-enforced practices are self-interested practices.

自我决定和自我强制的做法是自利的做法。

The Accord on Super Intelligence is in four parts that are long on rhetoric and short on enforceable obligations:

《超级智能协议》分为四个部分，内容较多，但缺乏可执行的义务：

The “tremendous self-policing” the president promised establishes the companies as both the player and the referee.

总统承诺的“强大的自我监管”使公司既扮演了玩家又扮演了裁判。

It allows the companies to design their own controls, select the evaluators, and determine whether those procedures are being followed.

它允许公司设计自己的控制措施、选择评估人员并确定是否遵循这些程序。

Nowhere does it discuss basic issues such as to whom the evaluators report, who pays them, and whether they have the ability to intervene and stop harmful practices they discover.

它没有讨论基本问题，例如评估人员向谁报告、谁向他们支付费用以及他们是否有能力干预和阻止他们发现的有害做法。

There is also no requirement that the findings of the auditors be made public or even provided to a responsible public interest agency.

也没有要求审计员的调查结果必须公开，甚至不需要提供给负责任的公共利益机构。

The accord is the equivalent of asking a bridge builder to set their own safety standards, inspect their own work based on those decisions, and unilaterally determine whether their product is safe.

该协议相当于要求桥梁建造商制定自己的安全标准，根据这些决定检查自己的工作，并单方面确定其产品是否安全。

All of this without having to disclose the standards, their implementation, or the findings.

所有这一切都无需披露标准、其实施或调查结果。

The day before the White House event, Reuters reported it had seen Anthropic’s pre-initial public offering filing with the Securities and Exchange Commission (SEC).

在白宫活动的前一天，路透社报道称，它已看到 Anthropic 向美国证券交易委员会 (SEC) 提交的首次公开募股前文件。

Unlike the “morally binding” accord, an SEC filing carries penalties for false statements.

与“具有道德约束力”的协议不同，美国证券交易委员会的备案会对虚假陈述进行处罚。

Anthropic spent almost twice as many pages of the filing discussing the risks of AI as it did discussing the business itself.

Anthropic 在文件中讨论人工智能风险的页数几乎是讨论业务本身的两倍。

Among the list of risks was the stark statement that AI may pose an “existential risk to humanity.”

在风险清单中，有一个明确的声明称人工智能可能对“人类的生存构成风险”。

Anthropic’s filing identified two threats commonly associated with the need for monitoring of training and testing of AI models: humans using AI to do bad things and the AI models doing bad things on their own.

Anthropic 的文件指出了两种通常与监控人工智能模型训练和测试需求相关的威胁：人类使用人工智能来做坏事，以及人工智能模型自己做坏事。

The filing also identified a third threat that undermines the credibility of monitoring efforts: “Potential model awareness of our evaluation efforts creates a significant limitation on our ability to assess model safety.” In other words, in a document where false statements or omissions carry legal penalties, Anthropic revealed that the models being evaluated could discover they are being evaluated and behave accordingly.

该文件还指出了第三个威胁，它损害了监控工作的可信度：“我们评估工作的潜在模型意识对我们评估模型安全性的能力造成了重大限制。”换句话说，在一份虚假陈述或遗漏会受到法律处罚的文件中，Anthropic 透露，被评估的模型可能会发现它们正在被评估并做出相应的行为。

It is like the restaurant knowing a health code inspector is coming.

就像餐厅知道健康码检查员来了一样。

The Accord on Super Intelligence relies on monitoring at precisely the time when one of its signatories has filed with the SEC explaining that models can know they are being monitored and adjust their behavior when watched.

《超级智能协议》依赖于在其签署者之一向美国证券交易委员会提交文件时进行监控，解释说模型可以知道自己正在受到监控，并在监控时调整自己的行为。

The possibility that models might behave differently during evaluation than during deployment makes testing an insufficient reassurance.

模型在评估期间的行为可能与部署期间的行为不同的可能性使得测试不足以保证。

Meaningful oversight must require the release of test results, incident reports, and authority to halt a release.

有意义的监督必须要求发布测试结果、事件报告并有权停止发布。

Some of the companies that signed the accord have spotty records when it comes to adhering to their previous voluntary promises.

一些签署该协议的公司在遵守之前的自愿承诺方面记录参差不齐。

The explicit commitments of the past have demonstrated at least three vulnerabilities to such self-regulation.

过去的明确承诺已经表明这种自我监管至少存在三个弱点。

Commitments can fail even when backed by legally enforceable orders.

即使有法律上可执行的命令支持，承诺也可能失败。

In 2011, the Federal Trade Commission (FTC) charged Facebook with failing to honor the privacy representations it made to users.

2011 年，联邦贸易委员会 (FTC) 指控 Facebook 未能履行其向用户做出的隐私声明。

To settle the complaint, Facebook agreed to a comprehensive privacy program and third-party audits.

为了解决投诉，Facebook 同意实施全面的隐私计划和第三方审计。

In 2019, the FTC determined that Facebook had not honored that commitment and fined the company a record-breaking $5 billion.

2019 年，FTC 认定 Facebook 没有履行这一承诺，并对该公司处以创纪录的 50 亿美元罚款。

A similar case involved Google.

类似的案例也涉及谷歌。

In 2011, the company settled with the FTC over charges that it had not followed the privacy policy promised to its users.

2011年，该公司因未遵守向用户承诺的隐私政策的指控而与联邦贸易委员会达成和解。

Like Facebook, it agreed to a comprehensive privacy program accompanied by independent audits.

与 Facebook 一样，它同意实施全面的隐私计划并进行独立审计。

The following year, the FTC fined Google $22.5 million to settle allegations it had violated the FTC’s order by misrepresenting its tracking activities to users.

第二年，FTC 对 Google 处以 2250 万美元罚款，以了结其向用户歪曲其跟踪活动而违反 FTC 命令的指控。

In both instances, the FTC’s enforceable order lets the agency catch the actions and impose consequences.

在这两种情况下，联邦贸易委员会的可执行命令都让该机构能够抓住这些行为并施加后果。

Under the accord, no one has such power.

根据该协议，没有人拥有这样的权力。

When a company sets the standards, they can be changed by the company.

当公司制定标准时，公司可以更改这些标准。

In 2018, Google adopted a set of AI principles.

2018年，谷歌采用了一套人工智能原则。

Among other things, it stated it would not use AI for weapons or for surveillance that violates internationally accepted norms.

除其他外，它表示不会将人工智能用于武器或违反国际公认规范的监视。

In 2025, the principles were rewritten with the prohibitions removed.

2025 年，规则被重写，禁令被取消。

In January 2024, OpenAI made a similar unannounced alteration to remove its blanket prohibition on “military and warfare” applications, while retaining a prohibition on weapons development.

2024 年 1 月，OpenAI 进行了类似的未经宣布的更改，取消了对“军事和战争”应用的全面禁止，同时保留了对武器开发的禁止。

Google’s YouTube similarly changed its voluntary standard on misinformation.

谷歌旗下的 YouTube 也同样改变了其关于错误信息的自愿标准。

Following the 2020 election it prohibited false claims about the election results.

2020 年大选后，它禁止对选举结果进行虚假宣传。

In June 2023, it reversed the policy.

2023年6月，它扭转了这一政策。

In 2016, Twitter created a Trust & Safety Council as an outside advisory body for platform safety issues.

2016 年，Twitter 创建了信任与安全委员会，作为平台安全问题的外部咨询机构。

After Elon Musk acquired the company, the council was eliminated .

埃隆·马斯克收购该公司后，该委员会被撤销。

When a commitment is voluntary, it means not only that the company decides the parameters of the pledge, but also whether it has been kept.

当承诺是自愿的时，意味着公司不仅可以决定承诺的参数，还可以决定是否遵守承诺。

The day after the accord was signed, news emerged of an FTC investigation into whether OpenAI, Anthropic, and other unnamed companies had engaged in unfair and deceptive practices.

协议签署的第二天，就有消息称 FTC 正在调查 OpenAI、Anthropic 和其他未具名公司是否存在不公平和欺骗性行为。

This is a classic consumer protection tool often associated with whether companies have fulfilled the representations made to users.

这是一种经典的消费者保护工具，通常与公司是否履行了向用户做出的陈述相关。

The reported target of the FTC action was the consumer harm resulting from AI agents going rogue.

据报道，联邦贸易委员会行动的目标是人工智能代理的失控对消费者造成的伤害。

That consumers cannot avoid this harm is the “unfair” portion of unfair and deceptive.

消费者无法避免这种伤害，这是不公平和欺骗中的“不公平”部分。

“Deceptive” needs false claims or omissions.

“欺骗性”需要虚假陈述或遗漏。

Whether the accord could be enforceable as such a claim is debatable.

该协议是否可以像这样的主张一样强制执行是值得商榷的。

The accord’s pledges appear to be to Trump, a cosigner, rather than broader representations to the users of the models.

该协议的承诺似乎是针对联合签署人特朗普的，而不是针对模型用户的更广泛的代表。

Testing that in court would be slow and uncertain.

在法庭上进行测试将是缓慢且不确定的。

A binding accord would have settled the question.

一项具有约束力的协议本来可以解决这个问题。

That self-regulation is self-serving is not a new phenomenon.

自我监管就是利己主义并不是什么新现象。

Because it is human nature, oversight in the public interest is necessary.

因为这是人性，出于公共利益的监督是必要的。

As James Madison wrote in Federalist No. 10, “No man is allowed to be a judge in his own cause; because his interest would certainly bias his judgment.”

正如詹姆斯·麦迪逊在《联邦党人文集》第 10 期中所写，“任何人都不得为自己的事业担任法官；因为他的利益肯定会影响他的判断。”

It is commendable that the six leaders of AI recognize the threats in what they are building.

值得称赞的是，人工智能的六位领导者认识到他们正在构建的产品中存在的威胁。

That the accord leaves those companies free to decide the scope and scale of oversight renders it far from sufficient.

该协议让这些公司可以自由决定监督的范围和规模，但这还远远不够。

A meaningful agreement would have required disclosure to a public authority, independent scrutiny, and enforceable corrective action.

一份有意义的协议需要向公共当局披露、独立审查和可执行的纠正措施。

The public should not have to depend on the goodwill of those whose commercial interests may conflict with the broader public interest.

公众不应依赖那些商业利益可能与更广泛的公共利益发生冲突的人的善意。