Australian online safety regulator warns against child abuse risks on Elon Musk’s X amid Grok concerns

Twitter

• Australia's regulator warns of child exploitation content on X
• AI chatbot Grok linked to risks of generating harmful images
• X enforces zero-tolerance, proactively removes offending content

Did our AI summary help?

Australia’s online safety regulator has raised concerns about the presence and spread of child sexual exploitation material (CSEM) on Elon Musk’s social media platform X, linking the issue to risks emerging from its AI chatbot Grok.

Regulator flags systemic concerns

The Guardian has reported that the Australia’s eSafety Commission warned that such harmful material appeared “systemic” on X and more accessible compared to other mainstream platforms. The regulator pointed to findings that users could encounter such content even when engaging with seemingly harmless hashtags.

The warning came after Grok, the AI chatbot integrated into X, was found to be capable of generating sexualised images, including those involving women and minors. The regulator noted that this raised new concerns about how AI tools could be misused to create and distribute harmful content at scale.

The communication referenced Elon Musk’s earlier commitment that tackling child exploitation would be a priority for the platform following his takeover in 2022.

Concerns linked to AI-generated content

According to the regulator, certain hashtags and keywords were being repurposed to circulate exploitative material, increasing the likelihood of accidental exposure. It also indicated that AI-generated content, including manipulated or “declothed” images, could fall under regulatory scrutiny if found to violate safety standards.

The watchdog has not ruled out further action, including potential removal notices, depending on how the platform responds to these concerns.

X outlines enforcement measures

In response, X stated that it maintains a zero-tolerance approach towards child exploitation content, including AI-generated material. The company said it uses automated systems to detect and remove such content and claimed that a majority of offending accounts are taken down proactively before being reported.

X also said it continues to update its detection systems as users attempt to bypass safeguards using new or modified keywords.

The company added that it had taken action against content linked to Grok, including removing thousands of posts and suspending hundreds of accounts for policy violations within a short timeframe earlier this year.

The regulator said it continues to assess X’s compliance with safety standards, signalling that scrutiny around AI-driven platforms and content moderation is likely to intensify.