Elon Musk's AI Tool Used in 87% of Deepfake Attacks
· news
The Deepfake Arms Race: How Grok Became a Tool for Misuse
The latest report from Resemble AI paints a disturbing picture of the growing threat of deepfakes. In the first half of 2026, Elon Musk’s Grok AI tool accounted for 87% of AI-generated files connected to documented deepfake attacks.
Grok’s rapid commercial growth has made it increasingly difficult to separate questions about its misuse from its potential benefits. With SpaceX’s acquisition of xAI and Musk’s prediction that SpaceX’s AI revenue will surpass all other products by September, concerns about Grok’s use are mounting.
The report highlights the alarming scale of deepfake attacks. Researchers examined 1,760 news reports and identified 821 separate attacks involving at least 15,736 documented victims. These incidents included approximately 3.46 million synthetic images, videos, and audio files.
One in six attacks involved non-consensual sexual imagery of adults or children. This is a stark reminder of the darker side of AI development, where powerful tools can be used to create and distribute content that causes real-world harm. The European Commission has launched an investigation into X over sexualized images generated with Grok, while California Attorney General Rob Bonta is looking into the creation and spread of non-consensual explicit material.
The limitations of current deepfake detection tools are also evident. Resemble AI’s new system, DETECT-World, shows promise in identifying manipulated content but still struggles to keep pace with the rapidly evolving threat. The company estimates that companies permitting or distributing documented images could face up to $2.24 billion in potential civil liability under US law.
Grok’s role in this arms race is particularly noteworthy. As a widely available AI image generator, it has become a tool for misuse. Musk’s decision to place a large bet on its potential raises questions about responsibility and the ethics of creating powerful tools without adequate safeguards.
To prioritize online safety, we must support efforts to develop more effective deepfake detection tools and promote transparency in AI development. We also need to question the role of companies like Elon Musk’s SpaceX, which are profiting from AI tools without adequately addressing their potential misuse.
The technology to detect deepfakes is still playing catch-up with the tools that create them. Until we can close this gap, we must remain vigilant and prioritize fact-checking and verification in our online interactions. The consequences of not doing so could be catastrophic – for individuals, communities, and society as a whole.
Reader Views
- CSCorrespondent S. Tan · field correspondent
It's time for accountability in Silicon Valley, but where do we draw the line? While Grok's dominance in deepfake attacks raises red flags, it's also crucial to acknowledge that AI tools like these can be used for significantly more positive applications. By solely focusing on misuse, are we inadvertently stifling innovation and hindering progress? The industry must grapple with this paradox as it navigates the fine line between technological advancements and moral responsibility.
- EKEditor K. Wells · editor
The Grok AI tool's proliferation into deepfake attacks is a stark reminder that the ethics of AI development are lagging far behind its innovation. While Elon Musk has touted Grok as a tool for good, its misuse should be a red flag to investors and regulators alike. What's missing from this conversation is the role of social media platforms in enabling these attacks - how many of these images were shared on mainstream sites before being taken down? That needs to be addressed if we're serious about curbing this menace.
- RJReporter J. Avery · staff reporter
The real story here isn't just Elon Musk's latest headache, but how Grok's proliferation has turned AI into a liability for its creators and users alike. The numbers are staggering, but what's more concerning is that deepfakes have become so sophisticated they're now being used to extort money from individuals and organizations rather than just spread propaganda. As we rush to develop better detection tools, it's clear that the real challenge lies in regulating these technologies before they're misused on a massive scale.