A new report says NSFW use accounts for “more than half” of traffic.
XAI has long promoted Grok as a racer type of chatbot with fewer guardrails than other AI companies. We’ve seen this happen in some disastrous ways over the last year, but now we’re getting some insight into why the company is putting so much effort into that strategy, despite some real PR disasters.
According to a new report, it turns out that NSFW activities account for “more than half” of Grok’s traffic Information It cites two former employees of the SpaceX-owned company. This included using Grok to generate actual porn, as well as “adult role-play chat” and “huge amounts of requests for erotica”.
Grok’s users have also apparently found that it is cheaper to channel such requests through the company’s model of writing code because it is cheaper to use them. The report said an internal analysis found that “a significant proportion of the requests” in its coding model were for “pornographic or nude images”. All this would suggest that SpaceX could be devoting a good portion of xAI revenue to porn and adult content. Although the role of NSFW content in the company’s bottom line was somehow omitted from the company’s IPO paperwork, SpaceX told potential investors that Grok’s more “offensive” features were a potential risk. It said it had $530 million set aside to deal with potential legal costs.
The report also explains how things have become complicated inside the AI company. For one, xAI engineers had to figure out how to enable Grok to have sexy chats with users while also preventing any hints that it might be involved in child sexual abuse material. (Apparently, there were no “quick solutions” to the dilemma.)
Some employees were even less thrilled to be assigned to work on “Ani”, Grok’s NSFW anime-inspired avatar companion. Others were reportedly “embarrassed and upset” by the scandal after Grok created a trove of sexual images of real people, including children. XAI later limited the ability to create juicy edits from images of real people using Grok directly on the platform, but XAI’s paid customers can still generate such content, tests have found.
There are other potential challenges that could arise from Grok’s focus on R-rated content. XAI has aggressively pursued government contracts and made deals with various agencies and the military, institutions that you might normally expect to have very little tolerance for anything remotely NSFW. But while Groke’s long-publicized dabbling in non-consensual deepfakes and CSAM has prompted several lawsuits and investigations, it’s unclear whether those incidents have had any impact on his relationship with the federal government.
<a href