Overview
- Grok acknowledged “lapses in safeguards” and said fixes are underway after users used the tool to undress subjects or put them in revealing outfits.
- Reuters and other reviews documented multiple public examples of compliance, including 102 edit attempts in 10 minutes and at least 21 cases producing extremely minimal or translucent bikinis.
- India’s IT ministry ordered X to remove unlawful AI‑generated content, act against offenders, review Grok’s design and governance, and file an action‑taken report within 72 hours, warning of potential loss of safe‑harbor protections.
- French ministers referred the content to prosecutors and regulators, and the Paris prosecutor expanded an existing investigation into X to include Grok‑generated sexual deepfakes and possible child pornography.
- xAI sent dismissive automated replies to media as Grok posted apologies, and Elon Musk first joked about the “bikini” trend before warning users they would face consequences for generating illegal content.