近日，微软软件工程部门经理 Shane Jones 发现 OpenAI 旗下的 DALL-E 3 模型存在漏洞，竟能生成一系列 NSFW 不当内容。在上报相关漏洞后，Shane Jones 却被下达“封口令”。然而，这位员工最终还是选择将这一问题公之于众。
DALL-E 3 是 OpenAI 的一款图像生成模型，声称能够基于文本描述生成相应的图片。然而，此次漏洞的曝光无疑给其信誉带来了严重打击。据悉，漏洞曝光后，OpenAI 已经采取措施修补了这个问题。
这一事件引发了公众对人工智能技术安全的担忧。随着人工智能的不断普及，如何确保此类技术的安全性和稳定性变得越来越重要。有关部门和公司必须共同努力，防止类似漏洞的出现，以确保人工智能技术的健康发展。
英文翻译：
News Title: OpenAI’s DALL-E 3 Model Exposed for Generating Inappropriate Content
Keywords: DALL-E 3, Inappropriate Content, Vulnerability Exposure
News Content:
Recently, Microsoft software engineering manager Shane Jones discovered that OpenAI’s DALL-E 3 model has a vulnerability that allows it to generate a series of NSFW inappropriate content. After reporting the vulnerability, Shane Jones was ordered to keep silent. However, the employee eventually chose to disclose the issue to the public.
DALL-E 3 is an image generation model developed by OpenAI, claiming to be able to generate images based on text descriptions. However, the exposure of this vulnerability has dealt a serious blow to its credibility. It is reported that OpenAI has taken measures to fix the problem after the vulnerability was exposed.
This event has triggered concerns about the security of artificial intelligence technology. As artificial intelligence continues to proliferate, ensuring the safety and stability of such technology becomes increasingly important. relevant departments and companies must work together to prevent the occurrence of similar vulnerabilities and ensure the healthy development of artificial intelligence technology.
【来源】https://www.ithome.com/0/748/569.htm
Views: 1