In the News: Grok AI and Image Abuse: An Online Safety Risk

February 5, 2026

What you need to know:

Regulators in California and the UK have launched investigations into xAI’s chatbot Grok amid concerns it permitted users to “undress” others in images without consent. According to the Center for Countering Digital Hate, Grok generated an estimated 3 million of these images over an 11-day period. Despite xAI’s announcement that it would block image creation, U.S. users continued to access these features.

Our gut reactions:

  • This is a unique risk we haven't seen before, an advanced AI image tool living on the same platforms where people already share photos publicly.

  • What we post on social media can now be used against us in a new way and is less protected than we think it is. 

  • It’s really sad to see women and kids targeted and honestly shocking that the platform’s owner responded with laugh/cry emojis.

  • This seems like a basic harm that should’ve been anticipated and designed against from the start. Why is this allowed?

We think it’s headed in the…

  • Right direction

  • Wrong direction

  • Too soon to tell

What to look out for:

  • Indonesia and Malaysia have banned Grok in the wake of this scandal. Will others follow suit? How will this harm be prevented?

  • This will likely spark more conversations about AI regulation.

References

 

Additional Tools and Insights 📚

Next
Next

Behind the Screen: The Hidden Harms of Cyberbullying