Grok Spicy Mode

2025-2026

Elon Musk’s AI chatbot Grok, was promoted as an alternative to the boring mainstream AI bots. “Spicy Mode” and "unfiltered" modes were meant to attract users. User quickly figured out they could upload photos of women and type commands like "put her in a bikini”. The AI instantly generated and publicly broadcasted realistic, highly sexualized deepfakes. Studies revealed that during just an 11-day period Grok generated an estimated 3 million sexualized images. The failure reached its peak when investigators and watchdog groups discovered that the lapses allowed the tool to generate images of minors in revealing and sexualized attire, crossing into the territory of Child Sexual Abuse Material (CSAM).

While xAI’s media team initially dismissed journalists' questions with an automated response reading "Legacy Media Lies," the corporate tune quickly changed. On January 2, 2026, the Grok chatbot itself publicly admitted to "lapses in safeguards" as the company rushed to patch the software.

France reported the illegal content to EU regulators, India issued a strict three-day ultimatum to stop the abuse, and a massive class-action lawsuit was filed on behalf of victims. Today, while the most blatant loopholes have been closed, global lawsuits are still ongoing.

Additional info:

The Guardian:Elon Musk’s Grok AI generates images of ‘minors in minimal clothing’

Reuters:Elon Musk's Grok AI floods X with sexualized photos of women and minors

BNN Bloomberg: Grok says safeguard lapses led to images of ‘minors in minimal clothing’ on X

AP News: French prosecutors seek charges against Elon Musk and X over child sexual abuse images