.png){kind=logo}
A French prosecutor has reached out to Elon Musk over serious allegations that X facilitated the spread of AI-generated deepfakes. This also raises doubts about the effectiveness of the platform's safety measures.
The latest development highlights the growing determination of European authorities to tackle tech firms, especially those operating social media platforms, over their responsibilities for AI governance, content regulation, and liability.
The French prosecutors took action following an investigation that uncovered a massive number of deepfake images available in bulk on the platform. They were worried about how easy it was to access the content, the inadequate moderation practices in place, and the use of Grok and similar software. As a result of raids conducted at their French offices, Elon Musk was called to testify before the prosecutors.
Below are some of the major accusations made against X (formerly Twitter) and Elon Musk:
Spread of explicit deepfake materials: Explicit sexual materials involving women shared on the platform.
Ineffective content moderation: Inability to delete harmful or illegal content on time.
Grok abuse: Possible misuse of this tool in creating and facilitating content.
Compliance issues: Potential regulatory issues with European digital rules.
The issues surrounding Grok emerged when there were reports of how the tool could generate pornographic material through deepfakes, based on the demands made by the users. There was no proper safeguard in place, making it common practice for people to use Grok to develop harmful tools. This made things worse for X, shifting the spotlight from its inability to censor user posts to the creation of harmful tools.
Also Read: Apple Threatens to Ban Grok Over Deepfake Abuse, Demands Major Safety Fixes
There are many critical factors that are involved in this process, and this can be illustrated through the example of France. In France, they are deciding whether the platforms will need to hold themselves accountable not just for the content created by users, but for the tools used by the users to create such offensive content.
