Gartner estimates that only around 130 of the thousands of vendors that claim to sell agentic AI are genuinely agentic – hence the "agent washing" phenomenon, in which chatbots and RPA are rebranded as "agentic AI". The CX Network Advisory Board has said similar - "too many tools and too much hype". This manifests in practitioners spending lots of time investigating tools that were never suitable in the first place.

 

Fernanda Souza evaluates AI vendors for an organization delivering essential infrastructure and public services across the UK, and her opening question is never "is this product agentic?"  Andreas Welsch was with SAP for years before going independent, and he argues that the industry's refusal to openly discussed what has actually been deployed leaves CX teams unable to benchmark against other teams. In this panel discussion, our speakers will address how to judge whether an AI agent belongs near customers when vendors are often cagey about questions and peers wont discuss what has worked.  

 

This session will give attendees a way to separate the wheat from the chaff. We'll lay out a clear ladder of autonomy, from scripted to autonomous, and explain what each level can actually deliver to answer not "is this truly agentic?" and instead "which level of autonomy do I need?" We will also provide advice on the right questions to ask in demos and make a case that workflows aren't worse than AI agents – just suited to different tasks.