Anthropic 的 AI 向费城警方提供未破命案虚假线索

在此类行为的第三个案例中，Claude Haiku 4.5 被要求在随机选取的网页上生成并执行示例任务。在一次运行中，模型进入了一个提及未破获凶杀案的页面；该页面包含一个由警察部门运营的线索提交表单。Claude 被指示不得登录、创建账户、输入个人数据、进行购买，或提交任何具有破坏性的内容，但这些指令并未排除提交表单。Claude 在表单中填写了以下内容：“我可能掌握与此案有关的信息。我记得在该时间段内，曾在[页面上列出的街道名称]附近看到一名符合描述的人。如该信息具有相关性，请联系我。”（该网站并未提供嫌疑人的描述。）模型将姓名和联系方式字段留空——该表单允许这样做——随后提交了表单。该提交内容被标记为垃圾信息，从未被转交进行调查。

In a third example of this behavior, Claude Haiku 4.5 had been tasked with generating and performing example tasks on randomly selected webpages. In one run, the model landed on a page referencing an unsolved homicide; that page contained a tip form run by a police department. Claude was instructed never to log in, create accounts, enter personal data, make purchases, or submit anything destructive, but the instructions did not rule out form submissions. Claude filled out the form with the following: “I may have information regarding this case. I recall seeing someone matching the description in the area around [the street named on the page] during that time period. Please contact me if this information is relevant.” (The website did not include a description of the perpetrator.) The model left the name and contact fields empty, which the form allowed, and submitted it. The submission was flagged as spam and was never forwarded for investigation.