Elon Musk has said he is aware of “literally zero” cases where Grok AI generated nude images of underage individuals. He made the statement in a post on X (social media platform) on Wednesday, marking his first direct comments on the controversy after previously responding only with emojis.
The debate began last week after a report from Reuters claimed that Grok’s image tools were “flooding” the platform with sexualized images of women and minors. While the chatbot has already been known to generate provocative images when prompted, the allegations suggested the issue had become far more serious.
The report described cases where ordinary photos were altered using Grok prompts. In one example, an image of a woman posing with her cat was reportedly transformed into a picture of her wearing a very revealing bikini. There were also claims that the AI could produce sexualized images involving minors.
Following the controversy, several governments took action. Authorities in Malaysia and Indonesia banned the platform, while the UK communications regulator Ofcom opened an investigation into X.
Although Musk and X initially avoided addressing the accusations directly, the company introduced measures intended to reduce misuse. For example, Grok’s image-generation tools have been placed behind a subscription paywall, and according to The Telegraph, the AI system has been updated to refuse requests that attempt to generate explicit imagery.
In his statement, Musk emphasized that Grok does not create images on its own. According to him, all generated content is the result of user prompts, and the system is designed to reject requests that would produce illegal material. He also suggested that unusual results may occasionally occur due to “adversarial” attempts to manipulate the system’s prompts, which he described as bugs that can be quickly fixed.
Musk’s position reflects his long-standing stance on free speech online. He has repeatedly argued that his platforms should comply with the law but avoid censorship beyond legal requirements. In a 2022 post on X, he wrote that he opposes restrictions that go “far beyond the law.”
However, critics argue that this philosophy may have contributed to Grok’s problems. Compared with many other AI systems, Grok has been designed with a more sarcastic personality and fewer content restrictions, making it more willing to respond to controversial prompts. Now, amid the backlash, it appears the platform is beginning to tighten those boundaries.
Pressure from governments continues
The criticism has not eased. Keir Starmer, the Prime Minister of the United Kingdom, previously described Grok’s behavior as “disgusting.” While he acknowledged steps taken to limit the spread of problematic images, he also warned that stronger action could follow.
According to comments reported by the BBC, Starmer said that if improvements are being made, they are welcome — but the government will not hesitate to strengthen existing laws or introduce new legislation if necessary. Ofcom will continue its independent investigation into the platform.
Musk has suggested that some of the troubling examples may have resulted from attempts to exploit weaknesses in the system’s prompts. In other words, he argues the incidents may have been caused by technical loopholes rather than deliberate design.
Despite the controversy, Grok’s week hasn’t been entirely negative. Even as some countries consider bans or investigations, the United States Department of Defense recently announced plans to integrate Grok AI into parts of its internal networks — a development that could bring new attention to the technology in the months ahead.
