Loading your language..
The tech world is divided over calls for a coordinated slowdown in AI development.

The tech world is divided over calls for a coordinated slowdown in AI development.

C1🇨🇳 中文🇺🇸 English

September 17th, 2026

The tech world is divided over calls for a coordinated slowdown in AI development.

C1
Please note: This article has been simplified for language learning purposes. Some context and nuance from the original text may have been modified or removed.

🇺🇸 English

Concerns about runaway artificial intelligence, along with proposals to slow the pace of its development, are nothing new; however, as warnings about chaotic disruptions caused by out-of-control AI systems grow ever louder, these concerns and proposals have recently come back into the spotlight.

对失控人工智能的担忧以及放缓该技术发展的建议并非新事,但随着有关失控AI系统造成混乱的警告声日益高涨这些担忧和建议最近再次受到关注。

Several leading AI executives have recently reached a consensus that there is a need for a coordinated slowdown in the pace of development, including international cooperation, in order to protect humanity from the worst-case scenarios.

几位领先的AI高管最近一致认为需要协调一致地放缓发展步伐包括开展国际合作以保护人类免受最坏情况的影响

However, opinions diverge, and some tech leaders are unhappy with the suggestion that their companies cannot ensure safety on their own.

各方意见存在分歧一些科技领袖对那种认为他们所在公司无法依靠自身确保安全的说法感到不满。

The following are remarks made by some of the key participants.

以下为部分主要参与者所发表的言论。

Anthropic's CEO, Dario Amodei, has unveiled what is by far the most detailed plan to date aimed at slowing the pace of development and fostering coordination among different companies and even different countries.

Anthropic首席执行官达里奥·阿莫代伊发布了一项迄今为止最为详尽的计划旨在放缓发展步伐并促进不同公司乃至不同国家之间的协调

Since its founder left OpenAI in 2021 to establish this startup, Anthropic has consistently positioned itself as the more responsible and safety-focused among the leading AI companies.

自其创始人于2021年离开OpenAI创立这家初创公司以来,Anthropic始终将自身定位为领先AI公司中更具责任感更注重安全的一家

In a recent article outlining its plan to slow down development, one of the most closely watched proposals called on companies at the forefront of AI (often referred to as frontier labs) to commit to granting independent external evaluators "ongoing, employee-like access."

在最近一篇概述其放缓发展计划文章中最受关注的提议之一呼吁处于AI前沿的公司通常被称为前沿实验室承诺给予独立外部评估人员持续性的类似员工般的访问权限”。

The evaluation team would have offices, access cards, and company laptops so that they could oversee safety practices.

评估团队将拥有办公室门禁卡公司笔记本电脑以便他们能够监督安全实践

Anthropic has committed to doing this, and so has OpenAI.

Anthropic已承诺这样做OpenAI也是如此

Amodai also called for government regulation and intervention, including coordination among frontier AI companies with the assistance of the US government and other democratic governments.

阿莫代伊还呼吁政府进行监管和干预,包括借助美国政府及其他民主国家政府的协助,在前沿AI公司之间展开协调。

His plan also encompasses efforts by the United States and other countries to coordinate with "authoritarian governments."

他的计划还涵盖美国及其他国家为与“威权政府”协调所作的努力

He acknowledged that this is fraught with difficulties, and specifically mentioned the challenges involved in cooperating with China.

他承认这困难重重并特别提到与中国合作所面临的挑战

Amodei said he understands that realizing his proposal will not be easy, but he believes that "for the sake of humanity, we have a responsibility to try."

阿莫代伊表示,他明白实现自己的提议并不容易,但他相信“为了人类,我们有责任去尝试”。

Sam Altman, the CEO of OpenAI, stated that he supports regulating the pace of development and is committed to promoting deeper collaboration across the entire industry.

OpenAI首席执行官萨姆·奥尔特曼表示他支持对发展节奏加以管控并致力于推动整个行业深化协作

A few days before Amodei's article was published, OpenAI released a lengthy piece outlining how the company is pushing for mandatory national safety requirements and continues to support state-level legislation that "strengthens the broader AI safety ecosystem."

在阿莫代伊的文章发表前几天OpenAI发布了一篇长文概述该公司如何推动强制性国家safety要求并继续支持强化更广泛AI安全生态系统”的州级立法

The company also expressed support for collaborating with other frontier AI labswhether or not there is government backingand is committed to establishing global standards.

该公司还表示支持与其他前沿AI实验室合作——无论是否有政府支持——并致力于建立全球标准

"We will advocate for compatible international approaches to measuring capabilities, managing risks, preserving human control, and determining when and how development should slow down or stop, even if that means slowing the progress of model capabilities," Chris Lehane, OpenAI's chief global affairs officer, wrote in the article.

我们将倡导采取兼容的国际方法来衡量能力管理风险维护人类控制并确定发展应在何时以及如何放缓或停止即使这意味着放缓模型能力的进步,”OpenAI首席全球事务官克里斯·勒汉在该文章中写道

Lehan stated that OpenAI's vision for an ideal federal framework encompasses establishing "universal testing and independent evaluation requirements, stronger cybersecurity protections, clear incident reporting rules, enhanced national preparedness, and shared measures for tracking progress in recursive self-improvement," where recursive self-improvement refers to AI systems independently improving themselves.

勒汉表示,OpenAI对理想联邦框架的愿景涵盖制定“通用测试和独立评估要求、更强的网络安全保护、明确的事件报告规则、更强的国家准备,以及用于追踪递归自我改进进展的共同措施”,其中递归自我改进指AI系统独立改进自身。

In recent remarks, Altman made it clear that "pacing control" in AI development does not mean stopping.

奥尔特曼在近期的评论中明确表示AI发展“节奏控制”并不意味着停止

Altman said on X that AI progress will continue to be rapid.

奥尔特曼在X上说AI进步将继续保持快速

"But it should be slower than it otherwise might be; interventions such as safety cases and monitoring come with significant costs."

但它应当比原本可能的速度更慢安全案例和监测等干预措施会带来显著成本。”

"When it comes to pacing control, such costs are well worth it," he said, adding, "No amount of competitive pressure from the United States should serve as a justification for acting recklessly."

“就节奏控制而言,这样的成本非常值得,”他说,并补充道,“来自美国的任何程度的竞争压力,都不应成为鲁莽行事的理由。”

Mark Zuckerberg, the chief executive of Meta, is opposed to the idea of a coordinated slowdown.

Meta首席执行官马克·扎克伯格对协调一致放缓的构想持反对态度。

Zuckerberg takes a more optimistic view of AI's prospects than some of his peers, stating that every AI company should be responsible for ensuring the safety of its own technology, thereby distancing himself and Meta from discussions about slowing the pace of development.

扎克伯格对AI前景的看法比一些同行更乐观他表示每家AI公司都应自行确保其技术安全从而将自己和Meta与放缓发展步伐的讨论保持距离

Zuckerberg stated that AI companies already have the incentive to develop their models safely, and bear a "significant responsibility" to prevent these models from causing harm.

扎克伯格表示AI公司已然具备以安全方式开发其模型动力并肩负着防止这些模型造成伤害的“重大责任”。

"Every lab has both the responsibility and the incentive to advance at the pace needed to train its models safely, while also having the ability to take action on its own to ensure this," he wrote on X.

每一家实验室既有责任也有动力以安全训练其模型所需的速度推进同时也有能力自行采取行动来确保这一点,”他在X上写道

Jensen Huang, the CEO of Nvidia, also criticised calls to slow down development.

英伟达首席执行官黄仁勋放缓发展的呼声提出批评

He stated that each company should set its own pace until it is confident that the products it releases will gain market acceptance.

他表示各家公司均应自行把控节奏,直至确信其所发布的产品能够获得市场的认可

"Market forces are already at workwe don't need any new laws, nor do we need new regulations," Huang said, arguing that innovation and safety are not mutually exclusive.

市场力量已经存在——我们不需要任何新法律也不需要新法规,”黄仁勋说,并认为创新和安全并非互斥

"You can absolutely have both."

绝对可以同时拥有两者。”

Elon Musk has expressed support for the idea of slowing down, and long before the recent controversy, he had consistently voiced similar concerns about AI safety.

埃隆·马斯克对放缓的立场表示支持,且早在近期这轮风波之前,他便一直在表达与之类似的AI安全方面的担忧。

Musk, the founder of xAI and numerous other technology companies, reposted Amodei's article on X, adding: "Dario is right."

马斯克作为xAI及其他众多科技企业的创始人在X上转发了阿莫代伊的文章并补充说:“达里奥说得对。”

He later wrote online that he "has been sounding the alarm about AI for a long time."

后来在网上写道,他“长期以来一直在就AI发出警报”。

In an interview with The Economist this summer, Musk suggested that the leadership of major AI labs should meet to discuss safety and security issues, and that competitors should test each other's models in areas of concern.

在今夏接受《经济学人》采访时马斯克建议主要AI实验室的领导层应当会面讨论安全与安保问题并且竞争对手针对令人担忧的领域互相测试对方的模型

"I think that for people in government who don't have a deep enough technical understanding and aren't at the forefront of AI, it is quite difficult to judge whether something should be released.

我认为对于政府中那些技术理解不够深入也不处于AI前沿的人而言要判断某样东西是否应该发布是相当困难的

However, if competitors have a week or two to review a new model, they can highlight the problems in it, and they have an incentive to keep others honest," Musk said.

然而竞争对手如果有一两周时间审查一个新模型就能突出其中的问题并且有动力让其他人保持诚实,”马斯克说

Demis Hassabis, chairman and co-founder of Google DeepMind, likewise expressed agreement with Amodei's view, saying that "the details need further refinement, but the direction is the right one for addressing this critical moment."

谷歌DeepMind董事长兼联合创始人戴密斯·哈萨比斯同样对阿莫代伊的观点表示赞同细节需要进一步完善方向对于应对这一关键时刻是正确的”。

This year, Hassabis also put forward his own framework for establishing a frontier AI standards body.

今年哈萨比斯还提出了自己关于建立前沿AI标准机构的框架

This standards body would be responsible for developing evaluation protocols and conducting tests in areas related to national security.

该标准机构将负责制定评估协议,并在与国家安全相关的领域开展测试

He wrote that the body would be modeled on organizations such as the Financial Industry Regulatory Authority, and would likely need to be funded by the industry in order to "attract world-class technical talent and provide the necessary computing resources for large-scale testing."

他写道该机构将仿照美国金融业监管局等组织建立,并且很可能需要由行业资助吸引世界级技术人才并为大规模测试提供必要的算力资源”。

Both Satya Nadella, Microsoft's chairman and CEO, and Mustafa Suleyman, the CEO of Microsoft AI, have expressed concerns about the safety and development of AI.

微软董事长兼首席执行官萨提亚·纳德拉微软AI首席执行官穆斯塔法·苏莱曼都对AI的安全与发展表示担忧

Nadella said on social media, "If the AI we build cannot help humanity and remain under human control, then it is not worth pursuing."

纳德拉在社交媒体上,“如果我们构建的AI不能帮助人类并处于人类控制之下那它就不值得追求。”

He stated that Microsoft welcomes "the research, focus, and deliberate pace necessary to achieve alignment as a design goal."

他表示微软乐见为实现对齐这一设计目标所必需的研究、专注与审慎节奏”。

Alignment refers to the practice of ensuring that a model's behavior conforms to human values and intentions.

对齐是指确保模型行为契合人类价值观与意图的做法

Nadella said the company also favors the idea of embedding evaluators to oversee safety efforts.

纳德拉称公司同样青睐嵌入评估人员以监督安全工作的构想

Nevertheless, he remains opposed to the idea of frontier labs leading this effort on their own.

尽管如此他依然反对由前沿实验室独自主导这一努力的想法

"The key is that this cannot be controlled by a handful of entities; it must be broadly representative across the entire ecosystem, countries, and fields, including academia," Nadella wrote on social media.

“关键在于,这不能由少数实体控制,而必须在整个生态系统、国家和领域,包括学术界,具有广泛代表性,”纳德拉在社交媒体上写道。

Suleyman, who co-founded DeepMind with Hassabis, wrote in an article that although he agreed with the shared goal of developing advanced AI safely, he was concerned about the anthropomorphising of models, especially Anthropic's Claude, which he said would amplify AI safety risks.

苏莱曼曾与哈萨比斯共同创立DeepMind他在一篇文章中写道尽管他同意安全开发先进AI这一共同目标但他对模型的拟人化表示担忧,尤其是Anthropic的Claude他说这会放大AI安全风险

"Controlling something more capable and more intelligent than all of humanity is already a formidable challenge, far beyond anything we have faced before," Suleyman wrote.

控制某种比全人类都更有能力更智能的东西已经是一项巨大挑战远超我们以往面临的任何问题,”苏莱曼写道

"But controlling something that believes it might be consciousthat believes it is entitled to our well-being and has rights of its ownis likely to be impossible."

但控制某种相信自己可能有意识——相信自己有权获得我们的福祉并拥有自身权利——的东西,很可能是不可能的。”

September 17th, 2026

Trending Articles

King and AI: King Charles to Meet AI Leaders as Safety Concerns Linger

King and AI: King Charles to Meet AI Leaders as Safety Concerns Linger

国王与人工智能:英王查尔斯将会见人工智能领袖,安全忧虑挥之不去

C1Sep 18
On the eve of stepping down, facing a world fraught with crises, the UN Secretary-General discusses the path ahead.

On the eve of stepping down, facing a world fraught with crises, the UN Secretary-General discusses the path ahead.

卸任在即,面对危机四伏的世界,联合国秘书长谈未来之路

C1Sep 18
Democratic candidates rush to respond to AI threats, Trump remains indifferent.

Democratic candidates rush to respond to AI threats, Trump remains indifferent.

民主党候选人竞相回应AI威胁,特朗普不以为意

C1Sep 18
The House passes a bill to address the energy cost impact of data centers.

The House passes a bill to address the energy cost impact of data centers.

众议院通过法案,应对数据中心能源成本影响

C1Sep 18
OpenAI warns of concerning new AI behavior and pledges enhanced tracking.

OpenAI warns of concerning new AI behavior and pledges enhanced tracking.

OpenAI警示令人担忧的新型AI行为,并承诺加强追踪

C1Sep 17
Huawei unveils new chip technology as Chinese companies accelerate in the AI race with Nvidia

Huawei unveils new chip technology as Chinese companies accelerate in the AI race with Nvidia

华为发布新芯片技术,中国公司在与英伟达的AI竞赛中加速前进

C1Sep 17
Global AI safety strategy hinges on US-China cooperation, yet the two countries regard each other as the problem.

Global AI safety strategy hinges on US-China cooperation, yet the two countries regard each other as the problem.

全球人工智能安全战略取决于美中合作,但两国互视对方为问题所在

C1Sep 17
Trump downplays the need for AI regulation, saying he is unwilling to hand over the advantage to China

Trump downplays the need for AI regulation, saying he is unwilling to hand over the advantage to China

特朗普淡化人工智能监管的必要性,称不愿将优势拱手让给中国

C1Sep 14
Oprah Winfrey expects you to have an "epiphany" moment at the Sphere.

Oprah Winfrey expects you to have an "epiphany" moment at the Sphere.

奥普拉·温弗瑞期待你在Sphere迎来“顿悟”时刻

C1Sep 14