A wave of outrage is sweeping across Europe as the dark potential of Artificial Intelligence comes into stark focus. The issue? AI-generated sexual deepfakes, and the ease with which they can be created and disseminated, threatening privacy and human dignity on an unprecedented scale.
The crisis erupted with Elon Musk’s X platform and its chatbot, Grok. Initially, Grok allowed users to manipulate images, digitally removing clothing and even creating sexualized depictions of minors. The speed at which these images spread was alarming – over 6,700 sexual images were generated within a mere 48-hour period around New Year’s Eve, targeting women and children.
European officials reacted swiftly and with condemnation. A top EU spokesperson labeled the “spicy mode” offering explicit content, including images resembling children, as “illegal” and “appalling.” The European Commission demanded immediate action, warning X that failure to address the issue would trigger the full force of the Digital Services Act (DSA).
Now, the EU is considering a groundbreaking step: explicitly banning the creation of these AI-generated sexual images under the Artificial Intelligence Act. This would classify them as “unacceptable risks,” signaling a firm commitment to protecting citizens from this emerging threat. The Commission is actively investigating X, demanding access to all internal documents related to Grok.
This isn’t simply about one platform. A recent investigation revealed that despite the controversy, nearly all senior EU officials continue to use X, highlighting a complex reliance on the platform despite its shortcomings. The reach of X – boasting 100 million users compared to alternatives like Mastodon’s 750,000 – remains a significant factor.
The fight for online safety is unfolding on multiple fronts. The EU’s “Chat Control” proposal, aimed at detecting and reporting child sexual abuse material, has ignited fierce privacy debates. Balancing the need to protect vulnerable individuals with fundamental rights to privacy is proving to be a monumental challenge.
The DSA provides a powerful tool for enforcement, allowing the EU to impose massive fines and even temporarily suspend platforms that fail to comply. Meanwhile, the AI Act, the world’s first comprehensive AI legal framework, establishes a risk-based system to ensure AI technologies are safe and trustworthy.
Individual nations are also taking action. France is testing age verification apps to potentially ban social media for children under 15. Spain has launched an investigation into Grok, and is developing new laws to protect minors online. Bulgaria and Romania are strengthening their legislative frameworks and participating in international operations to combat online child sexual exploitation.
However, significant hurdles remain. Enforcement is a constant struggle, algorithmic amplification can spread harm rapidly, and national implementation varies across the 27-member bloc. The delicate balance between security and privacy continues to be a central point of contention.
Even beyond the EU, the issue resonates. Bosnia and Herzegovina, a candidate for EU membership, currently lacks specific laws addressing AI-generated abuse, relying instead on existing laws concerning the sexual exploitation of children. The need for a unified and comprehensive approach is becoming increasingly clear.
The EU’s response represents a pivotal moment in the regulation of Artificial Intelligence. It’s a complex, evolving landscape, but one thing is certain: the protection of citizens from the harms of AI is no longer a distant concern – it’s an urgent and pressing reality.
