The trend of contact-free interactions and cost reduction has led to a steady increase in the use of KIOSKs, which, in turn, has increased the number of digital novices. To address this feature, various attempts, such as experience zones and the addition of voice features, have been made, but they have significant limitations which inhibit them from being the best solution. The limitation of experience zones is that they cannot cover all KIOSK screens. The addition of voice features can explain screen usage but cannot guide users on where to click for orders or cancellations in various scenarios. ChatGPT, a leading generative AI, continues to evolve and recently announced support for vision features. By utilizing this, we evaluated and tested the potential for a voice assistant more akin to a human helper, rather than simple voice guidance for screens, and obtained meaningful results.