Designing for AI Agents: UX Beyond Interfaces | Aayushman at UXINDIA25 скачать в хорошем качестве

Designing for AI Agents: UX Beyond Interfaces | Aayushman at UXINDIA25

AI is no longer just responding to commands — it’s planning, acting, and making decisions autonomously. So what does that mean for UX designers? In this forward-looking talk at UXINDIA25, Aayushman explores how UX must evolve for AI agents — shifting from designing static interfaces to designing relationships, trust, and collaboration between humans and autonomous systems. The talk breaks down the key shifts shaping agent-based experiences: From reactive tools to persistent, goal-driven agents From fixed user flows to dynamic systems From usability to alignment, transparency, and partnership Using clear examples (like AI travel planning and scheduling), Aayushman highlights common UX failures in agent systems — opacity, misalignment, loss of control, and trust gaps — and shows how better UX patterns can prevent them. This session is essential for designers building AI-powered products, copilots, assistants, and autonomous workflows. 🔥 Key Takeaways Why designing for AI agents means designing relationships, not screens The three major shifts: technology, behavior, and design UX challenges in agent systems: opacity, misalignment, trust gaps, and drift Designing for transparency: visible planning and explainability Goal negotiation and bounded autonomy in AI workflows UX patterns for agent systems: plan viewers, approvals, confidence signals, memory & feedback Why trust is built through user control, not just successful outcomes How to design effective human + AI partnerships