号外：欧盟就全球首个全面人工智能规则达成协议！（中英文对照）

LONDON (AP) — European Union negotiators clinched a deal Friday on the world’s first comprehensive artificial intelligence rules, paving the way for legal oversight of AI technology that has promised to transform everyday life and spurred warnings of existential dangers to humanity¹.

伦敦（美联社）- 欧盟谈判代表周五就全球首个全面的人工智能规则达成了协议，为人工智能技术的法律监督铺平了道路，该技术有望改变日常生活，并引发人类面临生存危险的警告¹。

Negotiators from the European Parliament and the bloc’s 27 member countries overcame big differences on controversial points including generative AI and police use of face recognition surveillance to sign a tentative political agreement for the Artificial Intelligence Act².

欧洲议会和该集团 27 个成员国的谈判代表克服了在生成式人工智能和警方使用人脸识别监控等争议问题上的巨大分歧，签署了《人工智能法案》² 的临时政治协议。

“Deal!” tweeted European Commissioner Thierry Breton just before midnight. “The EU becomes the very first continent to set clear rules for the use of AI.”

"成交！"午夜前，欧盟委员 Thierry Breton 在推特上写道。"欧盟成为第一个为使用人工智能制定明确规则的大陆"。

The result came after marathon closed-door talks this week, with the initial session lasting 22 hours before a second round kicked off Friday morning.

本周举行了马拉松式的闭门会谈，首轮会谈持续了 22 个小时，周五上午才开始第二轮会谈。

Officials were under the gun to secure a political victory³ for the flagship legislation. Civil society groups, however, gave it a cool reception as they wait for technical details that will need to be ironed out in the coming weeks. They said the deal didn’t go far enough in protecting people from harm caused by AI systems.

官员们为这项旗舰立法取得政治胜利的巨大压力而努力³。然而，民间社会团体却对该法案持冷淡态度，因为他们还在等待未来几周内需要解决的技术细节。他们表示，该协议在保护人们免受人工智能系统伤害方面做得还不够。

“Today’s political deal marks the beginning of important and necessary technical work on crucial details of the AI Act, which are still missing,” said Daniel Friedlaender, head of the European office of the Computer and Communications Industry Association, a tech industry lobby group.

计算机与通信行业协会（Computer and Communications Industry Association，科技行业游说团体）欧洲办事处负责人 Daniel Friedlaender 说："今天的政治协议标志着就人工智能法案的关键细节开展重要而必要的技术工作的开始，而这些细节仍然缺失"。

The EU took an early lead in the global race⁴ to draw up AI guardrails when it unveiled the first draft of its rulebook in 2021. The recent boom in generative AI, however, sent European officials scrambling to update a proposal poised to serve as a blueprint for the world.

欧盟于 2021 年公布了其规则手册的初稿，在全球范围内率先⁴ 制定了人工智能准则。然而，最近生成式人工智能的蓬勃发展，让欧洲官员争分夺秒地更新这项准备作为全球蓝图的提案。

The European Parliament will still need to vote on the act early next year, but with the deal done that’s a formality, Brando Benifei, an Italian lawmaker co-leading the body’s negotiating efforts, told The Associated Press late Friday.

共同领导欧洲议会谈判工作的意大利议员 Brando Benifei 周五晚些时候告诉美联社，欧洲议会仍需在明年初对该法案进行投票表决，但随着协议的达成，这只是一种形式。

“It’s very very good,” he said by text message after being asked if it included everything he wanted. “Obviously we had to accept some compromises but overall very good.” The eventual law wouldn’t fully take effect until 2025 at the earliest and threatens stiff financial penalties for violations of up to 35 million euros ($38 million) or 7% of a company’s global turnover.

在被问及这个结果是否包括了 Brando Benifei 想要的一切之后，他通过短信回复说："这非常非常好，显然我们不得不接受一些妥协，但总体来说非常好。” 最终的法律最早要到 2025 年才能完全生效，并威胁对违法行为处以严厉的经济处罚，最高可达 3,500 万欧元（3,800 万美元）或公司全球营业额的 7%。

Generative AI systems like OpenAI’s ChatGPT have exploded into the world’s consciousness, dazzling users with the ability to produce human-like text, photos and songs but raising fears about the risks the rapidly developing technology poses to jobs, privacy and copyright protection⁵ and even human life itself⁶.

像 OpenAI 的 ChatGPT 这样的生成式人工智能系统已经进入了世界的视野，它能够生成类似人类的文字、照片和歌曲，令用户眼花缭乱，但同时也引发了人们对快速发展的技术给就业、隐私和版权保护⁵，甚至人类生命本身⁶ 带来的风险的担忧。

Now, the U.S.⁷, U.K.⁸, China and global coalitions like the Group of 7 major democracies have jumped in with their own proposals to regulate AI, though they’re still catching up to Europe.

现在，美国⁷、英国⁸、中国以及七大民主国家集团（G7）等全球联盟都已提出了自己的人工智能监管建议方案，尽管它们仍在追赶欧洲。

Strong and comprehensive rules from the EU “can set a powerful example for many governments considering regulation,” said Anu Bradford, a Columbia Law School professor who’s an expert on EU law and digital regulation. Other countries “may not copy every provision but will likely emulate many aspects of it.”

美国哥伦比亚大学法学院教授、欧盟法律和数字监管专家 Anu Bradford 说，欧盟强有力而全面的规则 "可以为许多政府在考虑监管时树立一个强有力的榜样"。其他国家 "可能不会照搬每一条规定，但很可能会效仿其中的许多方面"。

AI companies subject to the EU’s rules will also likely extend some of those obligations outside the continent, she said. “After all, it is not efficient to re-train separate models for different markets,” she said.

她说，受欧盟规则约束的人工智能公司也可能会将其中的一些义务延伸到欧洲大陆以外。"她说："毕竟，为不同的市场重新培训不同的模型是没有效率的。

The AI Act was originally designed to mitigate the dangers from specific AI functions based on their level of risk, from low to unacceptable. But lawmakers pushed to expand it to foundation models, the advanced systems that underpin general-purpose AI services like ChatGPT and Google’s Bard chatbot⁹.

《人工智能法案》最初的目的是根据风险程度（从低到不可接受）来减轻特定人工智能功能带来的危险。但立法者推动将其扩大到基础模型，即支持通用人工智能服务（如 ChatGPT 和谷歌的 Bard 聊天机器人⁹）的高级系统。

Foundation models looked set to be one of the biggest sticking points for Europe. However, negotiators managed to reach a tentative compromise early in the talks, despite opposition led by France, which called instead for self-regulation to help homegrown European generative AI companies competing with big U.S rivals, including OpenAI’s backer Microsoft.

基础模型似乎将成为欧洲的最大症结点之一。尽管法国领导反对，要求进行自我监管，以帮助欧洲本土的生成式人工智能公司与美国的大型竞争对手竞争，包括 OpenAI 的支持者微软，不过，谈判代表们在谈判初期还是达成了暂时的妥协。

Also known as large language models, these systems are trained on vast troves of written works and images scraped off the internet. They give generative AI systems the ability to create something new¹⁰, unlike traditional AI, which processes data and completes tasks using predetermined rules.

这些系统也被称为大语言模型（LLM），是根据从互联网上搜罗的大量文字作品和图片训练而成的。它们赋予了生成式人工智能系统创造新事物的能力¹⁰，这与传统人工智能不同，传统人工智能使用预定规则处理数据和完成任务。

The companies building foundation models will have to draw up technical documentation, comply with EU copyright law and detail the content used for training. The most advanced foundation models that pose “systemic risks” will face extra scrutiny, including assessing and mitigating those risks, reporting serious incidents, putting cybersecurity measures in place and reporting their energy efficiency.

建造基础模型的公司必须编制技术文件，遵守欧盟版权法，并详细说明用于训练的内容。构成 "系统性风险 "的最先进基础模型将面临额外的审查，包括评估和降低这些风险、报告严重事故、采取网络安全措施以及报告其能源效率。

Researchers have warned that powerful foundation models, built by a handful of big tech companies, could be used to supercharge online disinformation and manipulation¹¹, cyberattacks or creation of bioweapons.

研究人员警告说，由少数几家大型科技公司建立的强大基础模型可能会被用于加强网络虚假信息和操纵¹¹、网络攻击或制造生物武器。

Rights groups also caution that the lack of transparency about data used to train the models poses risks to daily life because they act as basic structures for software developers building AI-powered services.

人权组织还警告说，用于训练模型的数据缺乏透明度，会给日常生活带来风险，因为这些模型是软件开发人员构建人工智能服务的基本结构。

What became the thorniest topic was AI-powered face recognition surveillance systems, and negotiators found a compromise after intensive bargaining.

最棘手的问题是人工智能驱动的人脸识别监控系统，谈判代表们经过激烈的讨价还价，最终达成了妥协。

European lawmakers wanted a full ban on public use of face scanning and other “remote biometric identification” systems because of privacy concerns. But governments of member countries succeeded in negotiating exemptions so law enforcement could use them to tackle serious crimes like child sexual exploitation or terrorist attacks.

出于隐私方面的考虑，欧洲立法者希望全面禁止在公共场合使用人脸扫描和其他 "远程生物识别 "系统。但各成员国政府通过谈判成功地获得了豁免权，这样执法部门就可以使用它们来应对儿童性剥削或恐怖袭击等严重犯罪。

Rights groups said they were concerned about the exemptions and other big loopholes in the AI Act, including the lack of protection for AI systems used in migration and border control, and the option for developers to opt out of having their systems classified as high-risk.

人权组织表示，他们对《人工智能法》中的豁免和其他重大漏洞感到担忧，包括对用于移民和边境控制的人工智能系统缺乏保护，以及开发者可以选择不将其系统归类为高风险系统。

“Whatever the victories may have been in these final negotiations, the fact remains that huge flaws will remain in this final text,” said Daniel Leufer, a senior policy analyst at the digital rights group Access Now.

数字版权组织 Access Now 的高级政策分析师 Daniel Leufer 说："无论最终谈判取得了怎样的胜利，事实上，最终的法案文本中仍将存在巨大的缺陷。

Tech reporter Matt O’Brien in Providence, Rhode Island, contributed to this report.

本报驻罗德岛普罗维登斯的科技记者 Matt O'Brien 对此报道有贡献。

KELVIN CHAN

Covering technology and innovation in Europe and beyond.

报道欧洲及其他地区的技术和创新。

References:

参考：

1.人类面临生存危险的警告

https://apnews.com/article/artificial-intelligence-risks-uk-summit-regulation-ce89e842a176e55e46b6cc493cd515a5

2.《人工智能法案》

https://apnews.com/article/ai-act-talks-artificial-intelligence-regulation-e2bb57eef7ba6d2b0c85ef757fcd3bb0

3.为这项旗舰立法取得政治胜利的巨大压力而努力

https://apnews.com/article/ai-act-artificial-intelligence-regulation-europe-06ab334caa97778770f5f57f4d904447

4.全球范围内率先

https://apnews.com/article/tech-ai-artificial-intelligence-europe-eu-15ac394679519084478e15217c156abc

5.版权保护

https://apnews.com/article/ai-copyright-artificial-intelligence-b456fa19dd90bea87533f97f043a31bc

6.人类生命本身

https://apnews.com/article/artificial-intelligence-risk-of-extinction-ai-54ea8aadc60d1503e5a65878219aad43

7.美国：拜登希望在人工智能保障方面快速采取行动，并签署一项行政命令来解决他的担忧

https://apnews.com/article/biden-ai-artificial-intelligence-executive-order-cb86162000d894f238f28ac029005059

8.英国：苏纳克说，英国峰会上达成的协议使人类在应对人工智能威胁的斗争中处于有利地位

https://apnews.com/article/artificial-intelligence-risks-uk-summit-kamala-harris-9e01b9d35ec1541755a273c07dfa4c71

9.谷歌的 Bard 聊天机器人

https://apnews.com/article/google-artificial-intelligence-bard-gmail-youtube-maps-1229638b82d19afb5226c913821fa1ad

10.创造新事物的能力

https://apnews.com/article/ai-hollywood-writers-strike-artificial-intelligence-dc71c4cabcca0ee1b1afe0050d392a36

11.网络虚假信息和操纵

https://apnews.com/article/artificial-intelligence-hamas-israel-misinformation-ai-gaza-a1bb303b637ffbbb9cbc3aa1e000db47

作者：KELVIN CHAN

翻译/校对：刘天栋 Ted

编辑：储黄蕾

相关阅读 | Related Reading

只有开源才能拯救AI