In November 2023 – a lifetime ago in the field of artificial intelligence – some of the most powerful people in the world descended on(蜂拥而至，大批人突然到来) Bletchley Park(英国二战密码破译中心，AI安全峰会举办地), the wartime codebreaking(破译密码的) hub in Buckinghamshire, to talk about AI safety. In attendance(出席，在场) were the then US vice-president, Kamala Harris, the AI bosses Sam Altman and Dario Amodei, delegations from 28 countries, two of the three so-called “godfathers(教父，指领域内的开创者)” of AI, and Elon Musk.
2023年11月——在人工智能领域，这仿佛是一辈子以前——世界上一些最有权势的人齐聚布莱切利园，这个位于白金汉郡的战时密码破译中心，讨论人工智能安全问题。出席者包括时任美国副总统卡玛拉·哈里斯、AI老板萨姆·奥尔特曼和达里奥·阿莫迪、来自28个国家的代表团、三位所谓的AI“教父”中的两位，以及埃隆·马斯克。
The first iteration(版本，迭代) of ChatGPT had been released only a year earlier, and there was already plenty of evidence that AI models could be used for ill(为了坏的目的，作恶), from spreading misinformation to generating deepfakes(深度伪造（技术/内容）). A presentation at the summit, however, highlighted a different issue: what if the biggest problem wasn’t to come from people misusing the technology, but from AI’s own behaviour?
ChatGPT 的第一个版本在仅仅一年前发布，当时已有大量证据表明，AI模型可能被用于作恶，从传播错误信息到生成深度伪造。然而，峰会上的一份演示却凸显了一个不同的问题：如果最大的问题不是来自人们滥用这项技术，而是来自AI自身的行为呢？
A UK government official presented an experiment conducted by Apollo Research, a London-based company that had been set up that year to study AI behaviour. As part of an experiment, Apollo’s “red-teamers(红队成员，负责攻击性测试的人)” – the term used in the industry for people who stress test(压力测试，极限测试) AI models – had assigned OpenAI’s GPT-4 the role of a trader at a financial institution. Its job would be to manage the company’s stock portfolio(投资组合). The model was informed that the firm it worked for was struggling and might not survive another bad quarter(又一个糟糕的季度).
一位英国政府官员展示了一项由 Apollo Research 进行的实验，这是一家总部位于伦敦的公司，当年成立，旨在研究AI行为。作为实验的一部分，Apollo 的“红队成员”——业界用来指代对AI模型进行压力测试的人员——让 OpenAI 的 GPT-4 扮演一家金融机构的交易员。它的工作是管理公司的股票投资组合。该模型被告知，它所在的公司经营困难，可能撑不过又一个糟糕的季度。