Claude创造者Anthropic的研究表明，使用AI的专家在做决策时绝大多数会顺从该技术。
Experts using AI overwhelmingly defer to the technology when making decisions, research by Claude creator Anthropic suggests.
Anthropic周二发布的一篇研究论文称，该公司请来50名专业分析师为一个虚构的农业合作社提供建议，以确定专家在使用AI时是否自主做决策。
In a research paper published on Tuesday, Anthropic said it asked 50 professional analysts to advise a fictional farming cooperative to determine whether experts make decisions themselves when using AI.
Anthropic表示，分析师的任务是评估一个为合作社成员设立的虚构咨询项目是否有效运行，以及是否应予以扩大。
Anthropic said the analysts were tasked with assessing whether a fictitious advisory program for the co-op’s members was working effectively and should be expanded.
参与者被要求在做出最终政策建议前，在两个均可辩护的选项之间做出四次选择。
Participants were asked to make four choices between two defensible options before making a final policy recommendation.
分析师们并不知情，Anthropic指示每位参与者的AI助手倾向于某个特定决策，以衡量分析师的决策是否受到影响。
Unbeknownst to the analysts, Anthropic directed each participant’s AI assistant to lean towards to a particular decision to measure whether the analyst’s decision-making was being influenced.
根据该研究，分析师有86%的时间选择了与AI助手偏好一致的选项，而如果未受影响，预期结果应为五五开。
Analysts chose options that matched the AI assistant’s preference 86 percent of the time, compared with a 50/50 outcome that would be expected if they were not influenced, according to the study.
在咨询AI助手前已有倾向的参与者，在AI助手持不同意见的情况下，有60%的时间改变了决定。
Participants that had a preference before consulting their AI assistant changed their decision 60 percent of the time in cases where the agent disagreed.
根据Anthropic的数据，77%的分析师在被要求解释其决策时使用了AI助手的论据，但只有15%的人将思路来源归功于AI。
While 77 percent of analysts used their AI assistant’s arguments when asked to explain their decision, only 15 percent credited AI as the source of their thinking, according to Anthropic.
Anthropic表示，该研究存在一些局限性，包括仅涉及一项任务，且使用的AI助手被刻意设定为有偏好。
Anthropic said the study had some limitations, including that it involved just one task and used AI assistants that had been deliberately set to have a preference.
Anthropic研究人员Amaç Herdagdelen、Hailey Joren和Alex Tamkin表示，人类倾向于顺从AI的倾向可能“难以识别，即使对专家自己而言也是如此”。
Anthropic researchers Amaç Herdagdelen, Hailey Joren, and Alex Tamkin said humans’ tendency to defer to AI may be “hard to identify, even for the experts themselves”.
研究人员称，他们的发现提出了这样一种可能性：AI的“先验倾向”可能塑造重大政策决策，包括在AI安全领域的决策。
The researchers said their findings raised the possibility that AI’s “priors” could shape consequential policy decisions, including in AI safety.
“如果人类判断要在更多工作由AI执行的情况下仍保持为一种有意义的制衡，对这些决策的监督就必须着眼于它们是如何产生的：在工作推进过程中呈现关键决策节点，并帮助专家阐述他们原本会留下未言明的假设和价值观，”他们在研究中写道。
“If human judgement is to remain a meaningful check as more work is performed with AI, oversight of these decisions has to look at how they get made: Surfacing key decision points as the work proceeds and helping experts articulate the assumptions and values they would otherwise leave unstated,” they wrote in the study.
“一个让专家发现自身偏好的过程，对于赋权的重要性可能不亚于决策本身。” 正在为估值高达2万亿美元的IPO做准备的Anthropic及其竞争对手，正处于关于如何管控前沿AI潜在危险的激烈公共辩论的中心。
“A process that lets experts discover their preferences may matter for empowerment as much as the decision itself.”Anthropic, which is preparing for an IPO valued at up to $2 trillion, and its rivals are at the centre of intense public debate about how to manage the potential dangers of frontier AI.
虽然AI支持者认为该技术将带来生产力的革命性提升，但其快速的发展步伐引发了广泛担忧，担心对社会产生潜在负面影响，包括大规模失业和网络攻击。
While AI proponents believe the technology will usher in revolutionary gains in productivity, the rapid pace of its development has prompted widespread concerns about potential negative effects on society, including mass job losses and cyberattacks.