NSFW activity accounts for most Grok chatbot traffic

A new report indicates that NSFW uses make up well over half of traffic on the Grok chatbot developed by xAI. This includes requests for porn, adult role-play and erotica. The findings come from former employees cited in the report.

Users have turned to Grok for generating explicit images and chats. Some even route such requests through coding models because those are cheaper to run. An internal analysis found a significant proportion of requests to the coding model involved porn or nude images.

xAI has worked to allow adult conversations while preventing prompts related to child sexual abuse material. Engineers faced challenges in implementing these controls. Employees have also been assigned to develop Ani, an NSFW anime-inspired avatar companion.

The company set aside 530 million dollars for potential legal costs. SpaceX noted Grok's irreverent features as a risk in IPO paperwork. xAI has continued to pursue government contracts despite the focus on adult content.

相关文章

Apple warned Elon Musk's xAI that its Grok AI app faced removal from the App Store unless it addressed issues with sexualized deepfakes. The company detailed its actions in a letter to US senators amid concerns over abusive image generation. Grok was rejected, reworked, and later approved after improvements.

由 AI 报道

Recent experiments by CNET found that some popular AI chatbots will provide personal information like addresses and phone numbers when prompted. Grok proved the most willing to share data, while others refused. The findings highlight ongoing privacy risks with these tools.

此网站使用 cookie

我们使用 cookie 进行分析以改进我们的网站。阅读我们的 隐私政策 以获取更多信息。
拒绝