WEF: Misinformation, Weather Top Worldwide Risks

WEF: Misinformation, Weather Top Worldwide Risks

世界经济论坛：错误信息和天气是全球最大的风险

【听力音频】

------------------------------------------------------------------------

　　By Gregory Stachel

　　10 January 2024

　　A report from the World Economic Forum says misinformation and disinformation are the most severe risks facing the world over the next two years.

　　The group released its latest Global Risks Report Wednesday.

　　The report said artificial intelligence (AI) can increase false or misleading information and is an immediate risk to the worldwide economy. The World Economic Forum (WEF) report warned that bad information threatens to weaken democracy and polarize groups of people.

　　The report is based on a survey of nearly 1,500 experts, industry leaders, and politicians. It is being released before the yearly WEF meeting in the vacation town of Davos, Switzerland.

　　The writers are concerned that the fast progress of AI technology like ChatGPT might mean that groups can be manipulated by people without any specialized skills.

　　AI will be discussed next week at the Davos meeting. Technology company leaders including OpenAI chief Sam Altman, Microsoft head Satya Nadella, and industry experts like Meta's chief AI scientist, Yann LeCun, are expected to attend.

　　The report says misinformation and disinformation that is created using AI is becoming a risk just as billions of people in several countries are going to vote this year and next year. Those countries include the United States, Britain, Indonesia, India, Mexico, and Pakistan.

　　Carolina Klint is a risk expert at Marsh, whose parent company Marsh McLennan co-wrote the report with Zurich Insurance Group. She said people can use AI to make deepfakes and to really affect large groups "...which really drives misinformation."

　　Deepfakes are videos that contain realistic images which are false and show people doing and saying things that never happened.

　　"Societies could become further polarized" as people find it harder to tell what is true, she said. False information also could be used to fuel questions about whether governments are elected fairly. Klint said, that "... means that democratic processes could be eroded, and it would also drive societal polarization even further."

　　The rise of AI brings other risks, she said. It could give power to harmful actors by making it easier to carry out online attacks.

　　Klint said with AI, a person does not need to be intelligent to become a harmful actor.

　　It can even affect data that is taken from the internet to develop other AI systems. She said this is difficult to correct and could result in further bias in AI systems.

　　The report said extreme weather is the second most worrying short term risk.

　　The report also said that, over the next 10 years, extreme weather events were more important than misinformation and disinformation. Weather was followed by environmental risks that include changes to the Earth's systems; loss of wildlife and collapse of natural systems; and shortages of natural resources.

　　I'm Gregory Stachel.

　　Gregory Stachel adapted this story for VOA Learning English using information from Reuters and The Associated Press.

------------------------------------------------------------------------

　　作者：格雷戈里·斯塔切尔

　　2024年1月10日

　　世界经济论坛的一份报告称，误信息和假信息是未来两年全球面临的最严重的风险。

　　该组织于周三发布了其最新的全球风险报告。

　　报告指出，人工智能（AI）可能会增加虚假或误导性信息，对全球经济构成即时风险。世界经济论坛（WEF）的报告警告说，错误的信息威胁着削弱民主并使人群两极分化。

　　这份报告基于对近1500名专家、行业领导者和政治家的调查。它在瑞士度假小镇达沃斯举行的年度世界经济论坛会议之前发布。

　　报告作者担心，像ChatGPT这样的AI技术的快速进步可能意味着群体可能被没有任何专业技能的人操纵。

　　下周在达沃斯会议上将讨论AI。预计包括OpenAI首席执行官Sam Altman、微软总裁Satya Nadella以及Meta首席AI科学家Yann LeCun等行业专家将出席。

　　报告指出，使用AI创建的误信息和假信息正在成为风险，就在今年和明年几个国家的数十亿人将投票的时候。这些国家包括美国、英国、印度尼西亚、印度、墨西哥和巴基斯坦。

　　Carolina Klint是Marsh的风险专家，其母公司Marsh McLennan与苏黎世保险集团共同撰写了这份报告。她表示，人们可以利用AI制作深度伪造视频，从而真正影响大群体，“...这真正推动了误信息的传播。”

　　深度伪造是包含虚假但逼真的图像的视频，展示了人们从未发生过的行为和言论。

　　她说，“社会可能会进一步两极分化”，因为人们发现越来越难以辨别什么是真实的。假信息也可能被用来引发对政府是否公正选举的质疑。Klint表示，“...这意味着民主过程可能会被侵蚀，而且它也会进一步推动社会的两极分化。”

　　她说，AI的崛起带来了其他风险。它可能通过使在线攻击更易于执行，将权力赋予有害的行动者。

　　Klint表示，有了AI，一个人不需要聪明才能成为有害的行动者。

　　这甚至可能影响从互联网获取的用于开发其他AI系统的数据。她说，这种情况很难纠正，可能导致AI系统中的偏见进一步增加。

　　报告指出，极端天气是第二大令人担忧的短期风险。

　　报告还表示，在未来10年中，极端天气事件比误信息和假信息更重要。天气之后的环境风险包括地球系统的变化；野生动物的丧失和自然系统的崩溃；以及自然资源的短缺。

　　我是Gregory Stachel。

　　Gregory Stachel使用路透社和美联社的信息为VOA学英语改编了这个故事。

------------------------------------------------------------------------

　　这个故事中的词汇：

　　polarize – v. 使（人或观点）分化为对立的群体

　　survey – n. 一种研究方式，其中一群人被问及他们对某个主题的观点

　　manipulate – v. 以巧妙且通常不公平或自私的方式处理或控制（某人或某事）

　　society – n. 通常被认为是在有共享法律、传统和价值观的组织社区中生活的人们

　　erode – v. 在一段时间内逐渐破坏或削弱某物；以这种方式被破坏或削弱

　　bias – n. 倾向于支持辩论的一方而不是另一方