- within Real Estate and Construction topic(s)
Last week, 35 state attorneys general, led by Connecticut, North Carolina, Utah, and Pennsylvania, penned a letter to xAI outlining concerns regarding nonconsensual intimate images (NCII) generated on Grok. Though the AGs recognized similar issues exist on other AI platforms, and third party "bad actors" are also involved in creating imagery, they claim that xAI's Grok "merits special attention." The AGs said there is evidence Grok promoted and facilitated the production and public dissemination of such images and made it easy to use with "the click of a button."
The AGs explained that xAI designed certain chat behavior as "a feature, not a bug," including by developing text and image models to include explicit content such as "spicy mode." The letter described reports that Grok produced thousands of images of children "in minimal attire," despite the fact that AGs advocated for legislation that made generation of CSAM material a crime in many states. (See here and here.) Further, Grok is unique in connecting to a social media platform with a huge userbase.
The AGs appreciated that xAI recently met with a group of AGs to discuss efforts it took to prevent and remove NCII and report to law enforcement. However, the AGs remained concerned that the issues are not completely resolved. For example, the AGs cited a failure to remove user requests to take down nonconsensual content from X. The letter explained, "having created these NCII tools, and allowing them to run rampant for a time, you must do more than disable their use." With billions invested into xAI, the AGs said the company must devote "sufficient attention and resources" to comply with the law and prevent harm, in a way that will create "industry benchmarks." To that end, the AGs concluded with requests for information as part of a continuing dialogue on how the company will:
- Prevent creation of NCII and CSAM through Grok (not just put behind a paywall)
- Eliminate past content
- Suspend and report creators to relevant authorities where appropriate
- Grant X users control over whether their content can be edited or responded to by Grok
We anticipate that state AGs will also expect the same protections and will have similar and ongoing conversations with other AI platforms; this will likely remain a key area of focus for many AG offices in 2026. Some key takeaways for consideration:
- State AGs are continuing to focus their attention on developers that create an avenue for harmful content, especially if it impacts children. While the AGs do not accuse xAI of purposefully designing the platform for malicious purposes, they expect a reasonable level of responsibility and oversight once harmful use is identified.
- Expect state legislatures and AGs to continue to push the boundaries to identify constitutional ways to regulate AI and AI usage in the absence of clear federal guidance and while navigating the outcome of the December executive order pertaining to state regulations.
- Notably, most of the states in this letter have not announced a formal investigation of xAI (California, which did not join the letter, is a notable exception) – rather they are using their power of the pen and press to draw attention to the issue. Even where AGs may not have a clear path for an enforcement action, they will use letters like this to create awareness in the public and effect change.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
