Enterprise Leaders Detail AI Agent Scaling Strategies
Business leaders from Citi, Ford, and Experian shared their approaches to scaling AI agents at Fortune's Brainstorm Tech conference. The discussions highlighted the critical need for trust, visibility, and control as foundational elements for deploying AI agents across enterprise operations. Companies are implementing centralized frameworks, robust monitoring, and clear governance to manage AI systems effectively, accelerating deployment while maintaining security.
Laura Heisman, Chief Marketing Officer of Dynatrace, emphasized that trust is central to current AI rollouts in business. She highlighted the common concern across industries: "Can you trust it? Is it right? And if it's wrong, can you stop it?" Heisman, along with other business leaders at Fortune's Brainstorm Tech conference, stated that building visibility and control into systems is essential for fostering this trust as AI agents increasingly chain together tasks.
Nikhil Joshi, Chief Information Officer in the markets division at Citi, detailed the financial giant's strategy for AI agent deployment. Citi has established a centralized technological foundation for all its applications and agents. This framework ensures that every agent is registered, monitored, audited, and governed, allowing for a single, consistent deployment process. Joshi noted that while this approach might seem conservative, it ultimately enables faster long-term scaling by building inherent trust.
Kathleen Peters, Chief Innovation Officer at Experian, described how the consumer credit reporting firm manages its deployed AI agents. Experian's system tracks the provenance of each agent, identifies the human employee who created it, and specifies the permissions each agent has to access data or perform tasks. Peters stated that this level of ecosystem understanding builds the trust necessary for rapid scaling.
At Ford Motor Company, Sammy Omari, Executive Director, Advanced Driver Assist Systems and In-Vehicle Infotainment, explained how AI is used to accelerate parts of the design-to-production process and facilitate a "fail fast" approach. Ford implements guardrails to manage this process. For instance, non-engineering employees can use AI-powered "vibecoding" tools to quickly generate computer code for new car features as initial proofs of concept. If an idea proves viable, engineers then write the final code from scratch, maintaining rigorous quality assurance before a product ships to consumers.
According to Fortune, these strategies underscore a shared industry focus on secure and governed AI deployment.
