The Perils of Agency: How Developers Perceive, Prioritize, and Address Risks in Agentic AI Products
For AI developers and product managers, it reveals a capability vs. risk control tension in agentic AI development, highlighting the need for better risk mitigation tools.
This study of 35 industry developers found that they perceive risks in agentic AI products as tied to agentic qualities like autonomy and tool use, prioritize product/business risks over societal ones, and lack mature controls for mitigating agentic risks without constraining functionality.
Agentic AI systems act autonomously, use tools, adapt to context, and operate in complex real-world environments. However, these same characteristics can create or exacerbate product risks. We studied how industry developers (n=35) perceive, prioritize, and address the risks in their agentic AI products. We found that developers' perceptions of risk were closely tied to the qualities that made the product agentic, such as autonomy, tool use, and usage in a real-world context. Developers prioritized product and business risks before considering downstream societal risks like job displacement and end-user privacy. This prioritization also impacted developers' ability and motivation to mitigate agentic risks. Finally, developers lacked mature controls for containing agentic risks, often relying on constraining the same characteristics that make agents useful: e.g., autonomy and goal complexity. These findings reveal a capability vs. risk control tension in agentic AI development: developers need to address risks that emerge from agentic capabilities, yet they currently have limited support for doing so without constraining agentic functionality.