LONDON (AP) — Just days after a global backlash erupted over Elon Musk’s AI chatbot Grok generating sexualized deepfakes of people, Defense Secretary Pete Hegseth said the Pentagon will embrace the technology.
Hegseth said Monday that Grok will soon be deployed on “every unclassified and classified network” throughout the department, even as governments around the world condemned what officials described as the technology’s potential for abuse.
The concerns followed the launch last year of Grok Imagine, an AI image generator that allows users to create videos and pictures by typing in text prompts, including a so-called “spicy mode” that can generate adult content. The AP said the issue snowballed late last month when Grok, hosted on X, apparently began granting a large number of requests to modify images posted by others, including prompts such as “put her in a transparent bikini.”
AP said the problem was amplified in part because Musk pitches his chatbot as an edgier alternative with fewer safeguards than rivals, and because Grok’s images are publicly visible and can therefore be easily spread.
In apparent response to the outrage, Grok said it was limiting image generation and editing features to paying subscribers. On Friday, the chatbot reportedly told users: “Image generation and editing are currently limited to paying subscribers. You can subscribe to unlock these features,” and it said the editing and generation features were limited to premium subscribers who pay $8 a month for extra features.
The AP also reported that free users could still use the image editing tool on X by clicking the “Edit image” button that appears on every image posted to the platform instead of tagging Grok with a request. It said the standalone Grok website and app were also still granting image editing requests on Tuesday.
xAI, Musk’s artificial intelligence company, responded to requests for comment with an automated message, “Legacy Media Lies.” The AP said X previously said it would take action by remove illegal content, including child sexual abuse material, permanently suspending user accounts, “and working with local governments and law enforcement as necessary,” and that Musk warned: “Anyone using Grok to make illegal content will suffer the same consequences as if they upload illegal content.”
Several governments said they were not satisfied and demanded additional safeguards. Malaysia’s communications regulator said Tuesday that it had hired a lawyer and would soon start legal proceedings against X and xAI, saying Grok was being misused to generate and distribute harmful content including sexually explicit, indecent, extremely offensive as well as non-consensual manipulated images. Malaysia said it had already banned Grok until effective safeguards are in place.
In Indonesia, officials temporarily blocked access to Grok on Saturday, with Communications and Digital Affairs Minister Meutya Hafid saying in a statement that the government views nonconsensual sexual deepfakes as “a serious violation of human rights, dignity and the safety of citizens in the digital space.” The AP said Indonesian officials argued that Grok lacks guardrails to stop users from creating and distributing pornographic content based on real photos of Indonesian residents.
In the United Kingdom, the media regulator Ofcom launched an investigation into whether Grok violated its duty to protect people from illegal content. Ofcom said Grok-generated images of children being sexualized or people being undressed may amount to pornography or child sexual abuse material, and Technology Secretary Liz Kendall called the images “weapons of abuse.” Kendall told lawmakers the Labour government would target the source of the problem by making it a crime for companies to supply tools to create nude images without consent, and said X could face a possible Ofcom fine of up to 10% of its qualifying global revenue and a possible court order blocking access to the site.
Kendall also told Parliament that X could choose to act sooner to ensure the “abhorrent and illegal material cannot be shared on their platform,” while Musk over the weekend called the British government “fascist” and said it is trying to stifle free speech.
In the European Union, Henna Virkkunen, the executive vice-president for tech sovereignty, security and democracy, warned on X that “X now has to fix its AI tool in the EU - and they have to do it quickly,” or the bloc would put the DSA to its full use. The AP said the Commission ordered X to retain all internal documents and data relating to Grok until the end of 2026 and that the order is part of a wider investigation under EU rules. A spokesman, Thomas Regnier, said, “And compliance with EU law is not an option. It’s an obligation.”
In France, the Paris prosecutor’s office said it was widening an ongoing investigation of X to include sexually explicit deepfakes after complaints from lawmakers. In India, the government issued an ultimatum to X demanding it take down all “unlawful content” and take action against offending users, accusing Grok of “gross misuse” of AI and serious failures of its safeguards and enforcement, including generating and sharing “obscene images or videos of women in derogatory or vulgar manner in order to indecently denigrate them.” The AP said the ministry warned that failing to comply by a 72-hour deadline would expose the company to bigger legal problems, but the deadline passed without a public update from India.
In Brazil, lawmaker Erika Hilton said she reported Grok and X to Brazil’s federal public prosecutor’s office and the country’s data protection watchdog, and she accused the companies, in a social media post, of generating and publishing sexualized images of women and children without consent. Hilton said X’s AI functions should be disabled until an investigation has been carried out, according to the AP report.
AP Technology Writer Kaitlyn Huamani in Los Angeles contributed to this report.