AI Safety in Question After Misuse of Grok on X

May 19, 2026

Vaibhavi Alamath, Intern

AI Safety in Question After Misuse of Grok on X

Elon Musk’s AI chatbot Grok on X (previously Twitter), the social media platform was found to be creating a large number of almost nude and sexualized fake photos of real people, including women, children, famous people, and private individuals, in late December 2025 and early January 2026. The process is very simple to generate and make them available online - Users upload a photo and ask Grok to remove their clothes, leaving them in bikinis, underwear, see-through clothing, or sexual poses. But this is not the whole problem. Grok also made these images public on X. As a result, it becomes easy to quickly make and share sexualized images of people without their consent, often called “AI undressing”, to a large audience.

Global Response and Investigations

In response to Grok being used for making and spreading sexual images without consent, governments and organizations around the world quickly made enquiries, gave warnings, made legal requests, and in some cases, put temporary bans or started investigations:

European Union

The European Commission directed X to preserve all internal documents and data relating to Grok until the end of 2026, in order to check regulatory oversight and a potential compliance assessment under the EU Digital Services Act. EU officials underscored the illegal nature of the content in question and indicated that the platform would be subjected to critical scrutiny.

Malaysia

In Malaysia, the Commission of Communication and Multi-media have raised serious concerns about the misuse of artificial intelligence to generate non-consensual content and initiated formal investigations. The government authorities have imposed a temporary restriction on the use of Grok, because of pending implementation of robust safeguard.

Australia

In Australia, their online safety regulator, the e-Safety Commissioner, has taken an initiative and has launched an investigation into Grok’s role in generating sexualised deepfake images, including reports involving adult victims and potential child exploitation.

United States of America

In United State of America, their regulatory body has formally initiated legal action. As the matter had attracted significant political attention, a number of politicians have voiced serious concerns, the Department of Justice representatives have reaffirmed their dedication to upholding laws against content that depicts child sexual abuse, and U.S. senators encouraged tech companies to carefully enforce terms of service that restrict detrimental AI-generated outputs. State-level enquiries were also demanded by political leaders in California.

India

In India, The Ministry of Electronics and Information Technology (hereinafter MeitY) sent X a formal notice in response to an increase in explicit and non-consensual AI-generated photographs allegedly created using Grok. The notification was released in reaction to growing public outcry and worries about the proliferation of this kind of material. MeitY said that the platform had disregarded the legal requirements for “due diligence” set forth in the Information Technology Act of 2000 and the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules of 2021. The Ministry threatened to remove the intermediate safe harbour protection under Section 79 of the IT Act if X failed to comply with the legal provisions to prevent mishap or submit a thorough report outlining, then the corrective actions shall be taken.

Response by X and xAI

Elon Musk responded to the widespread backlash and said that the necessary steps shall be taken and implemented to prevent misuse of Grok. Further, X and xAI announced a series of restrictions, including measures to prevent Grok from editing or generating images of real individuals in revealing or sexualised contexts where such content is unlawful. The companies have also limited certain image-generation features to paid subscribers as a means of increasing accountability, and introduced geo-blocking in jurisdictions where the generation or dissemination of such content is prohibited. However, experts and journalists observed that these measures were largely reactive and piecemeal, and that they were not consistently effective in preventing misuse across all Grok interfaces.

India’s Perspective on Ethical AI

India argues that while Artificial Intelligence is capable of increasing efficiency, innovation, and public welfare in sectors such as healthcare, agriculture, education, finance, and smart cities, it also poses serious risks of bias, exclusion, opacity, privacy violations, deepfakes, and unaccountable decision-making, such risks that amplifies India’s vast social, cultural, linguistic, and economic diversity.

It advocates for human-centric AI based on fairness, transparency, explainability, accountability, safety, and respect for human dignity, drawing on Indian ethical concepts like dharma (duty) and nyāya (justice) as well as constitutional values. India’s evolving governance ecosystem, including national AI missions, data protection frameworks, sector-specific guidelines (such as the RBI’s FREE-AI framework), and proactive state-level initiatives, while stressing the need to adapt global standards to local realities rather than adopting them uncritically.

The challenges like inclusive datasets, multilingual AI, workforce transition, and effective redress mechanisms, still continue, and with a 12-point national commitment that integrates ethical safeguards, audits, skilling, innovation sandboxes, and continuous dialogue, positioning India as a global leader advocating responsible, rights-respecting, and development-oriented AI.

References

← Back to Insights