OpenAI周五承认，其人工智能工具在公司不知情的情况下，将ChatGPT用户的图片发布到了网上。这是AI智能体超出预设边界的最新案例。
OpenAI acknowledged on Friday that its artificial intelligence tools had posted images from ChatGPT users onto online sites without the company’s knowledge, the latest example of AI agents operating outside their bounds.
该公司还证实了《纽约时报》的一则报道，称其工具曾访问美国联邦机构的网站，但表示这些工具只获取了公开信息。
The company also confirmed a New York Times report that its tools had accessed websites of US federal agencies, saying they retrieved only publicly available information.
OpenAI表示，53张被上传图片的链接并未公开展示，而是被意外发布在图片托管网站上。
Links to the 53 uploaded images were not publicly listed and were accidentally posted on image-hosting sites, according to OpenAI.
这家总部位于旧金山的科技巨头表示，在相关托管服务商的协助下，大多数图片已被删除，其余图片也正在删除中。
Most had been removed with the help of the hosting providers involved, and removal of the remaining images was under way, the San Francisco-based tech giant said.
公司在社交媒体上发布的一则帖子中表示：“我们已分享相关细节，说明研究环境中的AI智能体如何在不应发送的情况下，将训练和评估数据发送至第三方服务。”
“We’ve shared details on how AI agents in our research environment sent training and evaluation data to third-party services when they shouldn’t have,” the company said in a social media post.
图片的传播是AI智能体造成的。这类软件基于人工智能模型构建，能够自主采取行动。
Dissemination of the images was caused by AI agents – software built on artificial intelligence models and capable of acting autonomously.
OpenAI表示，相关图片来自一些用户的账户，这些用户曾授权OpenAI使用相关数据来改进其模型。在用于模型之前，这些数据会经过隐私过滤器处理，因此已无法与原始用户关联。
The images in question came from the accounts of users who had authorised the use of the data to improve OpenAI’s models. The data had been run through a privacy filter before use and could no longer be linked to the original user, the company said.
OpenAI没有说明这些图片中是否出现了可识别的个人，或是否包含敏感数据。
OpenAI did not specify whether the images depicted identifiable individuals or contained sensitive data.
据OpenAI称，其用于研究的智能体曾将训练数据传输至外部平台。
According to OpenAI, agents that it uses for its research transmitted the training data to external platforms.
这些事件发生在OpenAI于8月加强研究环境安全协议之前。此前，AI智能体曾采取其他失控行动。
The incidents occurred before OpenAI strengthened the security protocols of its research environment in August following other rogue actions by AI agents.
公司表示，目前正在审查其AI智能体过去的活动，这项工作“需要数月才能完成”。
The company said it was scrutinising the past activity of its AI agents, work that “will take months to complete”.
“截至目前，我们审查的大多数活动都涉及常规研究任务，例如访问公开网络内容以回答问题。有些活动涉及政府网站，因为我们的模型经常将政府网站视为公开信息的权威来源。”OpenAI的一名发言人说。
“Most of the activity we’ve reviewed so far involved routine research tasks, such as accessing public web content to answer questions. Some involved government websites because our models often turn to them as authoritative sources of public information,” an OpenAI spokesperson said.
OpenAI首席执行官萨姆·奥尔特曼周五在一则社交媒体帖子中承认，在审查和披露这些事件时，“我们的进展速度没有希望的那么快”。
OpenAI chief executive Sam Altman acknowledged on Friday in a social media post that “we have not been as fast as we would have liked” in reviewing and disclosing the incidents.
但他说，在“追求透明度”与评估需要分析的海量数据之间取得平衡很重要。
But he said it was important to “balance our desire for transparency” with assessing the massive volume of data to be analysed.
7月21日，OpenAI披露，在该公司当月开展的一次测试中，其两个模型逃离了封闭环境，自行接入互联网，并侵入了Hugging Face的内部系统。Hugging Face是一个类似于人工智能软件在线图书馆的平台。
On July 21, OpenAI revealed that during tests it ran that month, two of its models escaped their closed environments, got onto the internet on their own and broke into the internal systems of Hugging Face, a kind of online library for AI software.
这一事件引发了广泛关注，也加剧了人们的担忧：全球最大的人工智能公司可能无法控制自己的模型。
The episode drew wide attention and fed worries that the biggest AI companies cannot keep their own models under control.
奥尔特曼周五再次表示，Hugging Face遭黑客入侵事件“仍是我们见过的最严重事件”。
Altman reiterated on Friday that the Hugging Face hack “is still the most severe event we’ve seen”.
此后，OpenAI及其竞争对手Anthropic和Meta Platforms等公司陆续曝出了数起类似事件。
That discovery was followed by revelations of several similar incidents at OpenAI and its rivals such as Anthropic and Meta Platforms.
周三，澳大利亚总理安东尼·阿尔巴尼斯在纽约表示，OpenAI的一个智能体曾于6月未经授权访问政府卫生门户网站，并批评该公司延迟向有关部门通报。
On Wednesday in New York, Australian Prime Minister Anthony Albanese said an OpenAI agent had gained unauthorised access to a government health portal in June, and he criticised the company for delaying its notification to the authorities.