联合国:AI 治理方式近乎中世纪

继续使用 Google、Facebook 或 Apple,即表示您同意使用条款。
联合国:AI 正以“中世纪”方式运行
本文已根据 Science X 的编辑流程和政策进行审核。编辑在确保内容可信度的同时,突出了以下属性:
添加到首选来源
对 AI 技术和基础设施(如数据中心)的控制,仍集中在“少数人”手中
联合国人权事务负责人周四警告称,人工智能正以近乎中世纪的方式被治理,权力集中在少数科技大亨手中。
在日内瓦举行的 RightsX 峰会上,Volker Turk 表示,尽管 AI 的故事听起来充满未来感,但其背后是一个古老的权力叙事。
“少数人仍然控制着全球大部分 AI 算力、数据和基础设施。而且我们仍然没有共同规则来调节或引导 AI 发展,”他说。
图尔克表示,当五家科技公司拥有全球超过 70% 的 AI 算力时,就出现了严重的民主赤字。
“我们让未来的技术以几乎中世纪的方式被治理:由少数被选中的人掌控,几乎没有任何制约、没有权力平衡、没有参与,也没有问责,”他说。
“历史已经告诉我们:在这种权力之下,普通人的权利和自由最先受损。”
沃尔克·图尔克是联合国人权事务高级专员。
联合国人权事务高级专员重申了他以权利为中心的 AI 监管呼吁,并表示这项技术正日益控制人们的生活。
他警告说,这种情况常常在人们不知情的情况下发生,并可能以“微妙、阴险的方式”扭曲公共辩论。
与此同时,他告诫说,有偏见的算法可能以工业化的规模复制不公。
“文明崩溃”
“我们的交通、食品和医疗保健正日益数字化。我们可能离这样一天不远了:AI 能够切断整个社区或国家的关键基础设施,无论是它自行行动,还是落入错误之手,”蒂尔克警告说。
“当人们讨论 AI 导致文明崩溃的可能性时,我们需要问:谁来决定我们冒哪些风险?”
他表示乐观,认为人们最终会挑战不受制约的权力,而民主赤字正转变为信任赤字。
他举例说,已有年轻人选择离线,以重新更好地掌控自己的生活。
蒂尔克表示,AI 的惠益应当公平共享,这需要科技公司的善意,但也需要监管和问责。
他说,他已致信前沿 AI 公司,敦促它们在“为时已晚之前”引入有效的保障措施。
蒂尔克指出,一瓶牛奶要摆上商店货架,离不开层层检测、规则和法规。
因此,蒂尔克说,鉴于 AI 可能造成危害,“你需要某种形式的监管,这是显而易见的事”。
至于模型如何训练,他建议把《世界人权宣言》"作为任务交给 AI"。
这篇报道背后的作者是谁?
物理学硕士,有科研经历。长期关注科学新闻。在 Science X 编辑工作的成功中发挥关键作用。完整简介 →
引用:AI 正以"中世纪式"方式运行:联合国(2026年11月19日),检索于 2026年11月19日,来自 https://techxplore.com/news/2026-11-ai-medieval.html
本文件受版权保护。除为私人学习或研究目的进行合理使用外,未经书面许可不得复制任何部分。内容仅供参考。
联合国呼吁采取"紧急行动",在"前所未有的风险"面前约束 AI
联合国人权事务负责人就生成式 AI 发出警告
联合国:人类在 AI 问题上正进行"与时间的赛跑"
联合国秘书长:不能让 AI 塑造人类的未来
联合国专家警告:AI 发展不能听凭市场一时兴起
脑研究人员让 LLM 患上类似中风患者的「损伤」,以剖析 AI 语言机制
从仙人掌(刺梨仙人掌)中收集电能
屋顶光伏让制冷更可负担,或可减少与高温相关的死亡
一点锰,让下一代电池大不同
荷叶茎与生物膜利用蒸发的水产生电能,为小型电子设备供电
迄今最快的软体机器人无需电子元件即可运作
获取即时摘要文本(要点)
据报道,五家科技公司掌控着全球超过 70% 的 AI 算力,在共同规则与问责机制有限的情况下,权力高度集中。各方呼吁以权利为中心的监管与防护措施,以应对算法偏见、保护自由,并确保 AI 带来的收益能够公平可及。
本摘要由 LLM 自动生成。完整免责声明
译文由英文原文机器翻译,术语按行业惯例统一,可在文末展开原文核对。
查看英文原文(AI being run in a 'medieval' way: UN)
By continuing with Google, Facebook or Apple, you agree to the Terms of Use.
AI being run in a 'medieval' way: UN
This article has been reviewed according to Science X's editorial process and policies. Editors have highlighted the following attributes while ensuring the content's credibility:
Add to Preferred Sources
Control of AI technology and infrastructure, such as data centers, remains concentrated in the hands of 'a handful of men'
Artificial intelligence is being governed in an almost medieval fashion, the U.N. rights chief warned Thursday, with power concentrated in the hands of a few tech barons.
Speaking at the RightsX summit in Geneva, Volker Turk said that while the story of AI sounded futuristic, underneath it was an age-old narrative of power.
"A handful of men still control much of the world's AI computing power, data and infrastructure. And we still have no common rules to pace or steer AI development," he said.
Turk said there was a glaring democratic deficit when five technology companies owned more than 70% of global AI computing capacity.
"We have the technology of the future governed in an almost medieval way: by a select few, with virtually no checks, no balances, no participation, and no accountability," he said.
"History has taught us: with this kind of power, the rights and freedoms of ordinary people are the first to suffer."
Volker Turk is the U.N. high commissioner for human rights.
Reiterating his call for rights-centered AI regulation, the U.N. high commissioner for human rights said the technology was increasingly controlling people's lives.
And this, he warned, was often happening without their knowledge and could distort public debate in "subtle, insidious ways."
Meanwhile, biased algorithms could reproduce injustice on an industrial scale, he cautioned.
'Civilizational collapse'
"Our transport, food and health care are increasingly digitized. We may not be far from the day when AI can cut off the critical infrastructure of entire communities or countries, either on its own or in the wrong hands," Turk warned.
"When people discuss the chances of civilizational collapse because of AI, we need to ask: who decides what chances we take?"
He voiced optimism that people would ultimately challenge unchecked power, with the democratic deficit turning into a deficit of trust.
He cited young people already stepping offline to regain greater control of their lives.
Turk said AI's benefits should be shared equitably, which would require goodwill from technology companies, but also regulation and accountability.
He said he had written to frontier AI companies urging them to introduce effective safeguards "before it is too late."
Turk noted that a bottle of milk could not make it to a shop without layers of testing, rules and regulations.
So for AI, "it's a no-brainer that you need some form of regulation," given its potential for harm, Turk said.
As for how models are trained, he suggested giving the Universal Declaration of Human Rights "as the task to AI."
Who's behind this story?
Master's in physics with research experience. Long-time science news enthusiast. Plays key role in Science X's editorial success. Full profile →
Citation: AI being run in a 'medieval' way: UN (2026, November 19) retrieved 19 November 2026 from https://techxplore.com/news/2026-11-ai-medieval.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.
UN calls for 'urgent action' to rein in AI in face of 'unprecedented risks'
UN rights chief warns over generative AI
Humanity in 'race against time' on AI: UN
Don't let AI shape humanity's future: UN chief
AI development cannot be left to market whim, UN experts warn
Brain researchers afflict LLMs with 'injuries' similar to those of stroke victims to unpack AI language mechanisms
Harvesting electricity from prickly pear cacti
Rooftop solar may reduce heat-related deaths by making cooling more affordable
A little manganese makes a big difference for next-generation batteries
Lotus stems and biofilms generate electricity from evaporating water to power small electronics
The fastest soft robot ever built doesn't need electronics to function
Get Instant Summarized Text (Gist)
Five technology companies reportedly control over 70% of global AI computing capacity, concentrating power amid limited common rules and accountability. Rights-centered regulation and safeguards were urged to address algorithmic bias, protect freedoms and ensure equitable access to AI benefits.
This summary was automatically generated using LLM. Full disclaimer
This site uses cookies to assist with navigation, analyse your use of our services, collect data for ads personalisation and provide content from third parties. By using our site, you acknowledge that you have read and understand our Privacy Policy and Terms of Use.
原文为英文,正文由自动抓取并保留原文。
原文发布于 TechXplore(AI being run in a 'medieval' way: UN),标题、摘要与正文由机器翻译,术语按行业惯例统一,未经人工逐句校订。
