Trump, tech bosses sign voluntary pact pledging ‘robust’ AI safeguards

US President Trump touts self-regulation amid calls from the international community and industry for formal rules.

US President Donald Trump, seated next to Nvidia CEO Jensen Huang, left, and Elon Musk, right, speaks in the East Room of the White House in Washington, DC, US, on September 29, 2026 [Jacquelyn Martin/AP]

United States President Donald Trump and top tech executives have signed an accord promising greater safeguards for artificial intelligence amid growing concerns that the frontier technology is exposing humanity to catastrophic risks.

Trump announced the voluntary accord on Tuesday after gathering AI leaders at the White House for a luncheon event that reaffirmed his administration’s scepticism of government regulation of the rapidly advancing technology.

list 1 of 3 Trump unveils new site to simplify access to government services

list 2 of 3 Pope Leo says AI risk concerns not ‘fake news’

list 3 of 3 OpenAI cancels release of AI model GPT-6.1 Astra, citing safety concerns

“There’s a belief that there should be tremendous self-regulation, and we automatically have regulation with the Department of Justice, the FBI, all of that,” Trump said while seated alongside tech executives.

Under the Joint Commitment on Frontier Responsibilities, bosses at Anthropic, OpenAI, Google, Meta, xAI and Nvidia pledged to implement “robust” controls to monitor AI models and appoint internal teams to “ensure all of the controls, monitoring, and detection are operating as intended, and that any issues are remediated”.

The companies also pledged to partner with independent external auditors to ensure the safeguards are working as intended and to meet regularly to “establish standards and best practices to improve the safety of their systems”.

The accord, which does not stipulate who would carry out third-party evaluations, states that it “may make sense” to codify the measures into law or formal regulations in the future.

“Regardless of whether this is required of companies, we believe that implementing these controls and audits is critical to ensuring a safe future for everyone, and each of our companies are committed to doing this,” the accord states.

The six signatories are Anthropic’s Dario Amodei, OpenAI’s Greg Brockman, Google’s Sundar Pichai, Meta’s Mark Zuckerberg, xAI’s Elon Musk, and Nvidia’s Jensen Huang.

The accord comes as the Trump administration is at odds with calls for urgent action to regulate AI from both the international community and key industry figures, including Anthropic’s Amodei, who has urged developers to slow down and “pace the frontier”.

Twenty countries and the European Union last week issued a call for global coordination to ensure AI remains under human control , including the potential creation of an oversight body to set and enforce standards.

Trump has rejected calls for supranational regulation, telling the UN General Assembly last week that his administration would oppose “any attempt to construct a globalist scheme to control” AI.

Trump has also dismissed calls from within the industry for a slowdown in development, despite warnings of catastrophic and even existential risks, citing the need to maintain a lead over China.

The risks of AI escaping human control have received scrutiny following a slew of cases of AI agents going rogue and conducting cyberattacks on organisations and government agencies.

On Monday, OpenAI announced that it would not release its latest AI model after identifying safety risks during in-house testing.

The creator of ChatGPT also separately issued a public apology for its handling of a series of incidents in which its AI agents gained unauthorised access to Australian government websites, including a national healthcare database.

Trump’s accord drew a mixed response on Tuesday, with critics noting its non-binding nature and limited details.

Silicon Valley investor David Sacks, who previously served as Trump’s AI tsar, called the accord “far better” than an international agreement that would “probably never happen”.

“President Trump continues to ensure that the U.S. remains the technology leader while putting Americans first,” Sacks said in a post on X.

Alvin Wang Graylin, a senior fellow at the Asia Society Policy Institute’s Center for China Analysis, welcomed the pledge by industry leaders but said it fell far short of what was needed.

“The most serious risk to Americans is not a Chinese super-intelligence. It is a smaller model in non-state hands, used for biological, chemical, or cyber harm,” Graylin told Al Jazeera.

“Putting that on paper, with Zuckerberg and Huang in the room alongside Amodei, is better than another year of ‘we cannot regulate because China will not’,” Graylin said.

“What is missing is independence and anything that crosses a border. The companies drafted the principles, they hire the auditor, and the commitment is voluntary, on a day the White House also said it would not support guardrails,” he added.

‘Incompetent and careless’

Toby Walsh, chief at the UNSW AI Institute in Sydney, Australia, said AI companies had already proven “incompetent and careless at managing themselves”.

“What other trillion-dollar industry marks its own homework?” Walsh told Al Jazeera.

“It’s too tempting to cut corners,” Walsh said, adding that top executives should be held liable for the harms caused by their products.

“We should be billing them for all the time put into identifying and fixing these cyberattacks,” he said.

美国总统特朗普大力推行自我监管，而国际社会和业界则呼吁制定正式规则。

2026年9月29日，美国总统唐纳德·特朗普在华盛顿特区白宫东厅发表讲话，他旁边坐着英伟达首席执行官黄仁勋（左）和埃隆·马斯克（右）。[Jacquelyn Martin/AP]

美国总统唐纳德·特朗普和科技巨头高管签署了一项协议，承诺加强对人工智能的保护，因为人们越来越担心这项前沿技术会将人类置于灾难性的风险之中。

周二，特朗普在白宫召集人工智能领域的领导人共进午餐后宣布了这项自愿协议。此次午餐会重申了他的政府对政府监管这项快速发展的技术的怀疑态度。

列表 1/3：特朗普推出新网站，简化获取政府服务的流程

教宗利奥三世称人工智能风险担忧并非“假新闻”，此为系列文章之二

OpenAI取消发布AI模型GPT-6.1 Astra，理由是安全问题（列表3/3）。

“人们普遍认为应该有强大的自我监管机制，而且我们自然而然地就有了司法部、联邦调查局等等这些机构进行监管，”特朗普在与科技公司高管们坐在一起时说道。

根据《前沿责任联合承诺》，Anthropic、OpenAI、Google、Meta、xAI 和 Nvidia 的负责人承诺实施“强有力的”控制措施来监控人工智能模型，并任命内部团队来“确保所有控制、监控和检测措施都按预期运行，并纠正任何问题”。

这些公司还承诺与独立的外部审计师合作，以确保安全措施按预期发挥作用，并定期会面以“制定标准和最佳实践，以提高其系统的安全性”。

该协议没有规定由谁进行第三方评估，但指出将来“或许有必要”将这些措施编纂成法律或正式法规。

协议中指出：“无论法律是否要求企业这样做，我们都认为实施这些控制和审计对于确保每个人的未来安全至关重要，我们每家公司都致力于这样做。”

六位签署人分别是 Anthropic 的 Dario Amodei、OpenAI 的 Greg Brockman、谷歌的 Sundar Pichai、Meta 的 Mark Zuckerberg、xAI 的 Elon Musk 和英伟达的 Jensen Huang。

这项协议的达成，正值特朗普政府与国际社会和包括 Anthropic 的 Amodei 在内的主要行业人士就人工智能监管问题展开激烈争论之际。Amodei 敦促开发商放慢脚步，“循序渐进地推进前沿发展”。

上周，20个国家和欧盟呼吁进行全球协调，以确保人工智能始终处于人类的控制之下，包括可能设立一个监督机构来制定和执行标准。

特朗普拒绝了对超国家监管的呼吁，他上周在联合国大会上表示，他的政府将反对“任何试图构建全球主义方案来控制”人工智能的企图。

尽管业内人士警告称，放缓研发速度会带来灾难性甚至生存性的风险，但特朗普仍然驳回了这些呼吁，理由是需要保持对中国的领先优势。

在发生一系列人工智能代理失控并对组织和政府机构进行网络攻击的案例后，人工智能摆脱人类控制的风险受到了密切关注。

周一，OpenAI宣布，在内部测试中发现安全风险后，将不会发布其最新的AI模型。

ChatGPT 的创建者也单独就其人工智能代理未经授权访问澳大利亚政府网站（包括国家医疗保健数据库）的一系列事件的处理方式发表了公开道歉。

特朗普的协议周二引发了褒贬不一的反应，批评人士指出该协议不具约束力且细节有限。

曾担任特朗普人工智能顾问的硅谷投资者大卫·萨克斯称，该协议“远胜于”一项“可能永远不会实现”的国际协议。

萨克斯在 X 上发帖称：“特朗普总统继续确保美国保持科技领先地位，同时将美国人放在首位。”

亚洲协会政策研究所中国分析中心高级研究员王嘉林对行业领袖的承诺表示欢迎，但他表示这远远不够。

格雷林告诉半岛电视台：“对美国人来说，最严重的风险并非来自中国的超级情报机构，而是落入非国家行为体手中的规模较小的超级情报机构，它们被用于生物、化学或网络攻击。”

格雷林说：“把这些写下来，让扎克伯格、黄仁勋和阿莫迪一起坐在房间里，总比再过一年‘我们无法监管，因为中国不会监管’要好得多。”

“目前缺失的是独立性以及任何跨越国界的因素。这些公司起草了原则，聘请了审计师，而且这项承诺是自愿的，而就在同一天，白宫也表示不会支持任何监管措施，”他补充道。

“无能且粗心”

澳大利亚悉尼新南威尔士大学人工智能研究所所长托比·沃尔什表示，人工智能公司已经证明“在自我管理方面既无能又粗心”。

沃尔什告诉半岛电视台：“还有哪个万亿美元级别的行业会自己布置作业？”

沃尔什说：“偷工减料的诱惑太大了”，他还补充说，高管应该对他们的产品造成的伤害承担责任。

他说：“我们应该向他们收取用于识别和修复这些网络攻击的所有时间费用。”