Log in or create a free Rosenverse account to watch this video.
Log in Create free account100s of community videos are available to free members. Conference talks are generally available to Gold members.
When AI Agents Meet Reality. Service Design Lessons from a Pilot
Summary
Why does the introduction of AI agents so often lead to confusion instead of clarity? Drawing on an applied pilot project on Human-Agent Teams, this session shares key learnings from integrating AI agents into everyday work. The focus is not on the technology itself, but on structure. How roles shift. Where decisions get stuck. How trust is shaped or undermined. And why teams struggle even when the system technically works. The session shows how Service Design helped make these dynamics visible and manageable. Not as a creativity method, but as a way to design clear responsibility, decision paths, and collaboration between humans and agents. Presentation and discussion.
Key Insights
-
•
Agentic AI represents a shift from AI as a tool to AI as autonomous collaborators that run background workflows.
-
•
Many organizations have AI agents built, but a large gap remains in making them truly helpful and valuable.
-
•
Service design methods like process blueprints and human-in-the-loop approaches are key to designing effective AI-human workflows.
-
•
Building AI agents quickly is easy, but identifying where they add value requires deep problem understanding and organizational context.
-
•
Chaos in undocumented or inconsistent workflows complicates designing useful AI workflows.
-
•
Trust and responsibility must remain human-led with clear definitions of when and how humans intervene in AI processes.
-
•
Organizational power dynamics around information sharing critically impact the effectiveness of AI agent deployment.
-
•
Iterative quality feedback loops where AI agents improve based on continuous human input are vital for success.
-
•
Co-creative, multidisciplinary engagement including non-technical stakeholders increases adoption and reduces resistance to AI.
-
•
AI literacy at all organizational levels is necessary; leaders must engage with AI to properly support teams using it.
Notable Quotes
"Before for me, AI was just a tool, but now these agents are running in the background and are mostly invisible, changing everything."
"Lots of people have agents running, but they don't feel they’re really useful yet."
"It’s not about pure automation. It's like playing tennis—back and forth between human and agent."
"We must design who is in the loop and who is in the lead. The responsibility can’t be handed over to AI."
"If you spend more time building an agent than doing the task, it’s not worth it."
"The chaos of undocumented processes makes it hard to build useful AI workflows; mapping and blueprinting help."
"Information is power in organizations, so deploying AI agents means reconsidering who gets access to what data."
"It’s dangerous to believe everything AI says or to believe nothing. We must find a balance."
"You can't delegate AI decisions to IT alone; every team needs to figure out how AI agents support their work."
"If we don't do hard thinking and critical collaboration when building agents, we’ll end up with armies of unused agents."
Or choose a question:
More Videos
"Miles Davis gave the musicians the music as they entered the studio, and most first takes were the final ones."
Jim KalbachJazz Improvisation as a Model for Team Collaboration
November 6, 2017
"A really strong book has to be designed as a journey with a consistent voice guiding the reader."
Louis RosenfeldCoffee with Lou: Should You Write a (UX) Book?
March 7, 2024
"Levels can mean really different things at different companies, so ask a lot of questions about expectations for each level."
Catt Small Micah Bennett Brian Carr Jessica HarlleeWhat's Next for ICs: Exploring Staff and Principal Designer Roles
February 22, 2024
"You might find that spending the most time getting people interested in collaboration is where the work really happens."
Marieke McCloskeyUser Science: Product Analytics & User Research
March 11, 2021
"Replacing the entire body with an avatar future-proofs against unanticipated identifiers like tattoos or moles."
Llewyn Paine[Demo] Deploying AI doppelgangers to de-identify user research recordings
June 5, 2024
"Random assignment to treatment or control is essential—if it’s not random, it’s not a true experiment."
Joshua NobleCasual Inference
October 6, 2023
"Choosing the right output depends on the desired outcome—from quick iterations to gaining senior leadership buy-in."
Sara LogelYour Colleagues are Your Users Too
March 29, 2023
"Learning is a marathon, not a sprint, especially in the context of online conferences."
Bria Alexander Louis RosenfeldWelcome
January 8, 2024
"Consistency is so important that sometimes even consistency in failure works if it means I only have to learn the workaround once."
Sam ProulxOnline Shopping: Designing an Accessible Experience
June 7, 2023
Latest Books All books
Dig deeper with the Rosenbot
What unique challenges does healthcare research face when enabling honest participant feedback?
What is automation bias and how can UX design mitigate its risks in AI-augmented clinical decision support?
How can research enablement free up time for researchers to focus on high-level strategic questions?