Taming the Killer Robots With Humanity-Centered Design

Designing AI-infused products presents unique challenges, especially when it comes to creating human-centered experiences and applying Responsible AI (RAI) principles. Generative AI products, powered by Large Language Models (LLMs) and Multi-modal Models (MMMs), operate on probabilities and evolve through interactions, which contrasts with the deterministic nature of classical design and force us to design for being wrong.
The talk will explore how to navigate these complexities and design for uncertainty. Discover why the real threat isn’t killer robots, but “Botsplaining” models that confidently deliver incorrect information, potentially causing harm without users realizing it.
In the talk I’ll introduce the RAI-UX guidelines, a framework for designing AI-driven products that are human-centered, principled, proactive, collaborative, contextual, and holistic. The audience will learn what Responsible AI entails, its significance for user experience, and how UX professionals can integrate these principles into their work. I’ll also showcase real-world examples of how these guidelines have been successfully applied to LLM-based products at Microsoft.