Incredibly low prices, professional looking websites - but no delivery. Fake shops try to deceive consumers and tarnish the image of legitimate online retailers. eCommerce fraud reached an all-time high in 2019 (+32.3%). The Austrian reporting centre received more than 10,500 reports. However, important details of the ordering process (e.g. account details) are often missing and the fraud cannot be traced. This makes investigations difficult. 

With DETECT, interested users become valuable agents in the prevention and detection of crime in online retail through their involvement and use of the Fake-Shop Detector. Relevant authorities and stakeholders receive early warnings and can delegate tasks to the community.

Gamification is used to guide users through tasks. They are asked to accompany the review process by answering simple questions and documenting information. The resulting robust database contributes to an improved UX experience for all users, supports the work of the experts and strengthens the robustness of the AI models based on it.

A game to detect fake shops was successfully developed during the project. To support the artificial intelligence in the Fake Shop Detector (www.fakeshop.at), users are asked to check the automated assessments of the AI. In a playful way, they go on a journey through space and are slowly introduced to the world of fraud prevention experts by learning to distinguish fraudulent pirate camps (fake shops) from legitimate trading posts (reputable online shops).

Players are prepared for these challenges in 'home base training' sessions where they learn about trustmarks, secure payment methods, dodgy deals, online reviews and how the AI behind the Fake-Shop Detector works. Enlightened user involvement in the quality assurance process goes hand in hand with an introduction to safe online shopping and an explanation of how artificial intelligence works. In addition to supporting the quality assurance of the database by involving the community, skills in understanding AI are taught.