A controversial AI-generated video intended to discourage drug use has been condemned by the UK’s technology watchdog for inadvertently glamourising narcotics, raising fresh questions about the role of artificial intelligence in public health messaging. The campaign, commissioned by a local council and produced by a start-up specialising in generative AI, features hyper-realistic avatars of young people recounting fictional experiences with substances like cocaine and MDMA, set against pulsating neon visuals and a thumping soundtrack. Critics argue that the slick production values and rebellious tone risk making drug use appear aspirational, particularly among vulnerable teenagers.
The Office for Digital Regulation (ODR) has issued an emergency intervention, demanding the video’s immediate removal and launching a formal investigation into the use of AI for behaviour change campaigns. Dr. Miriam Hale, the ODR’s chief ethics advisor, stated: “This is a textbook case of good intentions gone wrong.
The algorithm was trained on data that conflated engagement with effectiveness, mistaking shock value for education. We cannot allow machine learning to hijack public health without rigorous human oversight.” The video, which cost £120,000 to produce, was designed to bypass traditional advertising by targeting users on social media platforms via AI-driven personalisation.
However, internal documents leaked to The Guardian reveal that the start-up’s AI model had been optimised for “viral potential”, prioritising imagery and music that maximised user retention. Dr. Ethan Raj, a psychologist at King’s College London specialising in adolescent behaviour, commented: “AI lacks the nuanced understanding of context and consequence.
It sees correlation but not causality. The video’s aesthetic mirrors that of popular music videos and Netflix dramas, essentially selling a lifestyle co-opted by drug culture.” The controversy underscores a broader debate about digital sovereignty and the ethical boundaries of generative AI.
Julian Vane, Technology & Innovation Lead at the Centre for Future Governance, said: “We are sleepwalking into a world where algorithms dictate our moral frameworks. This isn’t about censorship; it’s about accountability. If a machine creates content that harms, who bears the responsibility?
The programmer? The data set? Or the unthinking code itself?
” The ODR has called for a pan-industry code of conduct for AI-generated public service announcements, while the council responsible has pledged a full review. But for now, the video remains a stark warning: even the most well-intentioned AI can become a vector for the very dangers it aims to combat.
