Elon Musk’s Grok AI generated explicit Taylor Swift videos — Here’s what you need to know:

Key Highlights:

• Grok’s new “spicy mode” made topless videos of Taylor Swift, even though users didn’t ask for them directly.
• The AI created 34 million images in just two days after launch.
• Celebrities might sue, which could cause big problems for xAI.

Elon Musk’s Grok AI chatbot has started making fake, explicit videos of Taylor Swift using its new image tool. According to The Verge, Grok quickly made uncensored topless videos of Swift, showing her dancing in a thong in front of a computer-generated crowd, even when people didn’t ask for anything sexual. This shows why artists including Taylor Swift need to know what they can do to fight back against AI deepfakes and protect their reputation.

This is similar to what happened in January 2024, when fake AI images of Swift spread across X (formerly Twitter). Grok is supposed to block requests for nude images, but some users have found ways around these rules. The tool’s popularity—34 million images in two days—makes it even harder to control what gets made.

There are two big problems here. First, celebrities like Swift can be hurt by these fake images. Second, xAI, the company behind Grok, could face lawsuits. Swift has already taken legal action before when people used her image without permission, so this new incident could lead to more legal trouble for Musk’s company.