A coordinated global regulatory response to AI-generated deepfakes is emerging, with enforcement actions and legislative measures converging across multiple jurisdictions.
French prosecutors raided X’s Paris offices as part of an Europol-led probe into alleged unlawful data collection and complicity in child pornography possession, raising new scrutiny of the platform’s AI operations.
Thirty-seven state attorneys general are pursuing action against xAI over Grok’s non-consensual imagery generation capabilities, warning that the service may trample biometric privacy rights across the United States.
Stanford and Indiana University researchers found that Civitai, an AI-generated content marketplace, still hosts files designed to create deepfake pornography despite claiming to ban such material, reporting that 86% of deepfake requests involve LoRAs that adapt general-purpose models for disallowed outputs.
The U.S. Senate passed the Disrupt Explicit Forged Images and Non-Consensual Edits (DEFIANCE) Act in mid-January, and Illinois officials say the law must account for how training data may violate the state’s Biometric Information Privacy Act when scraped without consent.
International policy moves include Denmark’s new law regulating non-consensual synthetic media, and iProov reported an iOS video-injection tool that can turn apparently live footage into deepfake-style content, underscoring technical threats to biometric defenses.
The UK’s Information Commissioner’s Office and Ofcom both launched investigations into Grok’s image processing functionality, even as other watchdogs study the technology’s deployment.
Threat intelligence firm Group-IB warns that Deepfake-as-a-Service networks are fueling industrial-scale fraud, with biometric injection attacks sold on the dark web and delivered at scale.
Sources: Reuters (France raid), Reuters (state AGs), MIT Technology Review, The Hill, Reuters (Denmark law), iProov, The Verge, Group-IB
–
By Ali Nassar-Smith
