Unlike other mainstream AI photo-taking tools, Musk’s Grok appears to have fewer regulations. In CNN ’s test, for example, Grok easily created fake images of politicians, misleading voters.

Image of Elon Musk created by chatbox Grok. (Source: CNN)

Some X users also posted images created by Grok, showing celebrities smoking drugs, cartoon characters killing people, and sexualized images of women in bikinis.

According to CNN , Grok further raises concerns that AI could fuel an explosion of misinformation on the Internet, especially as the US presidential election approaches.

Lawmakers, activist groups and tech leaders have sounded the alarm that misuse of such tools could cause confusion and chaos for voters.

Many of the world's leading AI companies have implemented policies to prevent their AI-based image-generating tools from being used to create political misinformation, although researchers have found that users sometimes find ways around the rules.

Several companies like OpenAI, Meta, and Microsoft use technology or labels to help viewers know which images are generated by their AI.

Social networks including YouTube, TikTok, Instagram and Facebook also label AI-generated content in users' feeds, using technology to detect it themselves or asking users to identify when they posted it.

On August 16, AI appeared to impose some restrictions on Grok. The tool now refuses to generate images of political candidates or popular cartoon characters committing acts of violence.