UXtweak® reposted this
Could AI give unmoderated UX research more of a human touch? The idea of AI automation adding a "human touch" sounds like a paradox or even a joke. But when we’re talking about unmoderated UX research, which is already rigid and lacks real-time adaptability, it’s a valid question with both promise and pitfalls. Analyzing unmoderated UX research in the past 15 years, I often found myself wishing I could clone myself so that I could ask participants for clarification or keep them on track. While cloning myself remains in sci-fi, AI-powered moderation seems like a potential solution. Can AI improve unmoderated usability testing? Yes, but not without challenges. In my recent article for Smashing Magazine (link in the comments) based on a study by UXtweak® Research (also in comments), I discuss how generative AI has the potential to improve unmoderated usability testing. Yet, AI moderation also introduces challenges that can weaken feedback quality, such as: ♻️ Redundant and repetitive questions – AI might ask for clarification on something that’s already been answered or there’s nothing more to say, frustrating participants. 🤨 Off-topic or nonsensical questions - questions that no human in their right mind would ask in the context of the conversation. ⚖️ Methodologically unsound questions - bias can be introduced if not asking carefully 😡 Participant frustration - the more they interact with the AI, the more a negative bias grows. Furthermore, the backlash to AI oversaturation means there’s a good chance some participants will bear negative sentiment from the outset. Such issues need to be addressed before AI can assist with UX research moderation. What do you think about AI’s future in UX research? A tool for efficiency or a threat to insight quality?