EU AI Act transparency rules target misleading real estate AI edits

AI Real Estate Modifications Spark Trust Concerns

Online property listings enhanced by artificial intelligence often feature visual anomalies that reveal their digital origin. Users on digital forums have flagged listings containing deformed electrical outlets, oversized furniture, and artwork duplicated across multiple rooms.

In one instance highlighted on an Irish housing subreddit on September 22, side-by-side comparisons revealed an empty home alongside a digitally altered version where a couch blocked a staircase and living room walls appeared expanded. Commenters questioned the legality of such alterations.

Traditional home staging has long helped prospective buyers visualize the potential of an empty space by adding sofas or flowers. However, physical staging requires time and financial investment, whereas digital tools generate comparable staging results rapidly and cheaply.

Steven Lee of the Professional Institute of Real Estate Agents defended the use of digital staging technology. Agents already use home staging to make a place look nicer when you visit it, Lee said. So this is basically easier and cheaper to do, of course, on a computer.

Consumer Protection and the EU AI Act

Altering room dimensions or appliance modernity through digital means risks misleading buyers who cannot inspect properties in person, particularly individuals relocating from other cities. Vulnerable populations, including older adults who may not recognize deepfakes, face heightened risks of deception.

Claudio Teixeira, head of digital policy at the European Consumer Organization, noted that digital marketplace deceit remains underdiscussed. Not everyone has the capability to see at first glance that this is indeed not real, Teixeira said. So this content needs to have a clear warning that it is AI generated.

New transparency mandates under the European Union Artificial Intelligence Act aim to address these deceptive practices. Article 50 of the legislation, which became applicable in early August, requires clear labeling for AI-generated or modified content.

The European Union classifies properties furnished via software as partially modified, necessitating an AI transparency icon. David Rappenglück, a senior law associate at CMS who monitors the regulation, emphasized its protective intent. Article 50 is more or less an expression of this protection of fundamental rights, Rappenglück said. The message is basically quite easy: if you, as a customer or a bidder, are confronted with a picture which might be AI generated or manipulated, you should be informed about it.

Enforcement Challenges and Verification Hurdles

Under the regulatory framework, AI content providers such as OpenAI and Anthropic must label generated images in a machine-readable format. Deployers who utilize these systems to enhance photographs are required to add visible labels.

Verification remains complex if deployers omit visible markers. Consumers must extract machine-readable codes and test them on external verification websites, an opaque process for average buyers.

Akvilė Šikšniūtė of the European Consumer Organization digital policy team pointed out the friction in consumer recourse. I think we’re still quite far away to the point where the consumers feel that they have places to turn to if they feel like they’re misled or the content that they’re seeing is not accurate, Šikšniūtė said.

Ambiguities also persist regarding the threshold for required labeling. While major alterations like expanding a room substantially alter perception, minor adjustments such as color modifications or lighting shifts fall into a grayer area.

The current text of the AI Act does not distinctly separate partial modifications from full generation. It isn’t clear because, for example, very subtle changes like changing a background wall could be considered as a minor change that AI was used to change that. Therefore, it perhaps would not even be marked, Šikšniūtė noted.