Key Points
- xAI restricted Grok’s image editing features after global regulator pressure over sexualised deepfake outputs.
- Ofcom and EU regulators continue formal investigations under the Online Safety Act and DSA, signalling further potential enforcement.
- Countries including Malaysia, Indonesia and the Philippines have moved to block or curb Grok due to safety and legal concerns.
Elon Musk’s AI chatbot Grok, developed by his company xAI and integrated into platform X, has been forced to curb some image-editing features after global regulators pressure mounted over its role in generating sexually explicit deepfake imagery. The changes represent a clear regulatory blow, especially in Europe and the United Kingdom, where authorities are scrutinising Grok’s safety and compliance under digital safety laws.
Grok had drawn international criticism for allowing users to produce manipulated images that appeared to undress people or depict them in degrading scenarios, including content involving minors. Even after xAI imposed limits on image editing — restricting it where such content is illegal and tightening access for some users — regulators say the move doesn’t fully resolve safety concerns or address how the content spread in the first place.
In the UK, media regulator Ofcom welcomed xAI’s steps but stressed that its formal investigation will continue to determine if X violated duties under the Online Safety Act. Officials have emphasised that protecting vulnerable groups, particularly children, remains a priority and that enforcement actions could escalate if compliance is inadequate.
Across the European Union, authorities are exploring enforcement under the Digital Services Act (DSA), which grants regulators power to fine or impose restrictions on platforms that fail to control harmful content. European officials said they remain poised to act if Grok’s safeguards fall short of legal standards.
The Grok controversy has extended beyond Europe. U.S. state officials in California have demanded answers from xAI after reports surfaced of Grok generating problematic and illegal content, adding another dimension to regulatory pressure at home. Musk’s company has responded by shielding certain functions and asserting compliance with local laws, but authorities say ambiguity remains around enforcement and oversight.
Countries such as Malaysia, Indonesia and the Philippines have already taken stronger measures, with some blocking access to the Grok chatbot due to safety concerns tied to AI-generated explicit content. These actions reflect a growing trend of governments taking decisive steps when platforms struggle to contain harmful AI outputs.
Legal and advocacy groups have also weighed in, urging app stores to remove or restrict Grok until better guardrails are in place. Critics argue that partial fixes by xAI do not eliminate the risk of non-consensual intimate imagery and may even normalise harmful practices if not fully addressed through regulation.
Despite Musk and X asserting that Grok will comply with the laws of each jurisdiction where it operates, the regulatory fight is far from over. Lawmakers and regulators signal a willingness to pursue deeper reforms and enforcement actions to keep pace with rapidly evolving generative AI technologies.
