“警觉”
用英语怎么说？

When citizens cast their ballots, their votes are supposed to stay confidential. However, artificial intelligence now threatens this fundamental privacy. A Princeton University researcher tested whether AI could identify individual voters using publicly available election records and standard language models. Within hours, he built a workflow capable of analyzing records across Georgia. The AI agent easily processed the data without raising any ethical objections. This revelation has raised alarms among officials just before early voting begins in the upcoming midterm elections, exposing serious weaknesses in the protection of confidential votes.

当公民投下选票时，他们的选票理应保持保密。然而，人工智能现在对这种基本隐私构成了威胁。普林斯顿大学的一位研究人员测试了是否可以利用公开的选举记录和标准语言模型通过人工智能识别选民个人身份。在短短数小时内，他就建立了一个能够分析佐治亚州全州记录的工作流程。该人工智能代理在未提出任何伦理异议的情况下轻松处理了这些数据。就在即将举行的中期选举提前投票开始前夕，这一发现引发了官员们的警觉，暴露出在保护秘密选票方面的严重缺陷。