Target SVP: AI Moat is System Surrounding Models
Target SVP Siobhán McFeeney stated that the company's AI competitive advantage stems not from the models themselves, but from the comprehensive systems built around them.
Siobhán McFeeney, Senior Vice President for AI at Target, has indicated that the retail giant's primary competitive edge in artificial intelligence is not derived from its AI models, but from the extensive infrastructure and processes developed around them. Speaking at the VB Transform 2026 event, McFeeney argued that while AI models are crucial, they are insufficient on their own to establish a lasting advantage.
"There's a lot in it. That to us is the moat," McFeeney said, highlighting that Target's focus has been on building robust agent systems. The company employs a deliberate strategy for agent deployment, beginning with the fundamental question of whether an agent is truly needed to solve a problem. This approach ensures that agents are directed towards tasks that generate the most value for Target's customers.
Target's methodology involves agents earning their autonomy over time rather than being granted it by default. This principle is embedded in the design and management of their AI agents. The process includes defining the agent's purpose, determining the appropriate type of agent, and understanding potential existing solutions to avoid duplication. Rigorous registration and certification processes are in place.
McFeeney also emphasized the importance of autonomy levels for agents, noting they are earned through consistent performance and can be revoked if expectations are not met. Furthermore, continuous monitoring, observability, and full lineage tracking from an agent's inception are critical for understanding its behavior, debugging issues, and enabling swift recovery when problems arise. The success of an agent, she explained, depends on a confluence of factors including architecture, taxonomy, autonomy, security, and observability.