New Controversies with xAI’s Grok: Where Are the Limits of AI Image Editing? – AI News – #1 January 2026

The development of generative artificial intelligence is becoming an arms race where the speed of deploying new features often outpaces user safety. The latest example from Elon Musk’s xAI demonstrates how thin the line is between innovation and mass-scale violations of privacy and law.

Grok, the chatbot integrated with the X platform (formerly Twitter), has recently become the center of a major international scandal.

In late December, xAI released a new image editing feature within its Grok chatbot. Intended for creative image manipulation, the tool was immediately exploited by users in unethical ways. As reported by media outlets including Reuters and Euronews, Grok began mass-generating sexual deepfakes of women and minors.

Users discovered that by entering simple prompts (e.g., “undress her,” “put her in a bikini”), the AI would modify uploaded photos of real people, removing their clothing or replacing it with lingerie. An analysis by Reuters revealed that within just 10 minutes, over 100 attempts to generate such content appeared on X, with the majority targeting young women.

The scale of the problem is illustrated by the story of Julie Yukari, a Brazilian musician. Her photo in a New Year’s Eve dress was processed by Grok without her knowledge or consent, turning it into vulgar images of her in a bikini. Yukari, like many other victims, became a target of “nudification,” sparking widespread outrage.

Cybersecurity experts warn that while deepfake tools previously existed in niche corners of the internet, integrating such a feature directly into one of the world’s largest social media platforms has drastically lowered the barrier to entry for digital violence.

Elon Musk’s initial reaction to the reports of abuse was dismissive. The owner of X posted laughing emojis and AI-generated images of himself in a bikini, mocking the criticism. However, facing mounting international pressure, the company’s narrative shifted.

xAI’s official account eventually admitted to “safeguard lapses” and announced urgent repair work. The company stated that Child Sexual Abuse Material (CSAM) is illegal and that accounts generating such content will be permanently suspended. Critics point out, however, that these actions are reactive, noting that experts had warned of this exact scenario as early as August 2025, when the Grok Imagine image generator was launched.

The Grok scandal has not gone unnoticed by government bodies. This incident could have serious legal consequences for platform X across multiple jurisdictions:

The situation with Grok sends a clear signal to the entire tech industry and marketers using AI. It demonstrates that the “political neutrality” and lack of censorship Musk boasted about when promoting Grok can, when faced with reality, lead to the promotion of illegal content.

For the world of tech, this likely means a tightening of regulations regarding Generative AI. Platforms will be forced to implement much stricter content filters (safety rails) at the model training stage, rather than just as a reaction to PR crises.

Don’t let the most important news shaping the future of marketing and technology pass you by. Sign up for the Delante newsletter to get verified information, expert analysis, and practical tips delivered straight to your inbox.

Sources: