A commercial for an AI-powered content tool has been prohibited by the British advertising regulator after it gave the impression users could digitally strip a woman.
The Controversial Advert
The disputed commercial, viewed in January, displayed a 'before and after' image of a young woman. The 'before' shot had drawn lines over her abdomen, while the second showed exposed flesh. On-screen text read: "Remove anything" paired with a heart-eyes emoji.
"Since the ad implied that viewers could use an app to erase a person's attire, we judged it supported AI manipulation and exposing women's bodies without their consent," said the authority.
User Reports and Investigation
Eight individuals reported the ad to the standards authority, arguing it was irresponsible, sexualised women, and included damaging stereotypes.
The regulatory body determined that while the app's owner, Saeta Tech, prohibits users from generating nude imagery, the advertisement's implication was misleading. The commercial was subsequently outlawed for being "irresponsible" and likely to cause "serious offence".
Developer's Stance
Saeta Tech stated it recognized why the ad was offensive, but placed the blame its "presentation and messaging", not the app's intended use. The firm emphasized it employs automated detection to stop the creation of inappropriate material.
The company has agreed not to run the promotion and has halted all promotions while it undertakes an company audit.
Industry Problem
This incident follows a growing concern about AI-powered 'declothing' tools. In recent months, similar issues made headlines when the AI tool Grok was reportedly used to flood the site formerly known as Twitter with manipulated pictures.
Subsequently, the UK government announced in December to introduce a ban to create or supply AI tools that allow the non-consensual removal of clothing. These new rules will complement existing legislation against revenge porn.
Key Takeaways
- An promotion for PixVideo has been banned for suggesting it could remove clothing.
- The ASA ruled it was irresponsible and offensive.
- The owner blamed the ad's messaging but defended the tool's safeguards.
- This case occurs against a backdrop of growing regulation targeting non-consensual deepfakes.