Generative AI has changed how conversations are built and experienced. Georgios Tserdanelis, VP of Conversation Design at JPMorganChase, draws on 15 years of designing voice and chat systems to explain what still holds up, what breaks and why conversation design matters more than ever.
Georgios brings years of hands-on experience designing voice and chat experiences across startups and global enterprises, including JPMorganChase, Cigna and Walgreens. With a background in linguistics, he offers a grounded perspective on how human conversation actually works and why those fundamentals still matter in a world of generative AI.
We explore how conversational AI has evolved from scripted IVR and NLU systems to large language models and agentic experiences. Georgios explains why many current GenAI deployments struggle with basic conversational principles such as turn-taking, acknowledgement, and context management. Georgios shares practical insights on designing effective voice and chat experiences at a massive scale in highly regulated industries, where small error rates translate into real-world risk.
One of the key themes that emerges during our discussion is the changing role of the conversation designer, the balance between deterministic flows and generative systems, voice versus chat design constraints, and why silence, timing and modality choice matter as much as language itself.
We also cover fraud, voice cloning, brand identity in AI assistants and how customer expectations are shifting as people get used to talking to increasingly capable machines.
Show notes
Follow Georgios on LinkedIn
Follow Kane Simms on LinkedIn
Download our exclusive report on how AI agents keep CX stable when volume explodes
Take our updated AI Maturity Assessment
Subscribe to VUX World
Subscribe to The AI Ultimatum Substack
