Designers Face Challenges With AI Stop Button
Designers are struggling to create a stop button for AI agents, which can interpret and act on their own. This issue matters because it raises concerns about autonomy and control.
Original reporting · uxdesign.cc
Original: “Everyone is building AI agents. Who’s designing the stop button?”
Read “Everyone is building AI agents. Who’s designing the stop button?” at UX Collective →
How this was written. GDS wrote this article with a language model from UX Collective’s reporting (model: @cf/meta/llama-3.3-70b-instruct-fp8-fast); it describes that reporting rather than adding to it. The original, linked above, is the full story.
Key points from UX Collective’s report
- Designers are struggling to create a stop button for AI agents.
- The hardest design decisions happen where AI autonomy ends.
- Researchers are working to address the challenges of designing AI systems.
Designers are facing new challenges as AI agents become more autonomous, according to UX Collective. The traditional approach to designing predictable user flows is no longer sufficient, as AI can now interpret, decide, and act on its own. This shift in AI capabilities has significant implications for designers, who must now consider where the autonomy of AI agents ends.
The hardest design decisions are happening at the point where AI autonomy ends, UX Collective reported. This is because designers need to determine how to effectively stop or limit AI agents when necessary. The issue of showing where AI stops in the flow is a critical one, as it requires designers to think carefully about the boundaries of AI decision-making.
Researchers have been exploring this issue, including those at Thinking, who have been investigating the challenges of designing AI systems. In September 2025, researchers at Thinking were working on this problem, highlighting the ongoing nature of this challenge. The use of AI agents is becoming increasingly common, making the need for a well-designed stop button more pressing.
Questions and answers
What is the main challenge for designers with AI agents?
The main challenge is determining where the autonomy of AI agents ends and how to effectively stop or limit them when necessary.
Who is working on this problem?
Researchers, including those at Thinking, are working to address the challenges of designing AI systems.
