Enterprise Executives Detail AI Agent Governance and Deployment

Executives from John Hancock, Mercedes-Benz, and Estée Lauder discussed the operational execution and governance of AI agents within the enterprise. The panel emphasized a human-in-the-loop approach to manage high-risk, customer-facing decisions, particularly within the insurance and automotive sectors.

Karthik, Global CIO for John Hancock, advocated for strict governance and human-centric guardrails to ensure safety and explainability in regulated environments. Daniela Avanzini of Mercedes-Benz detailed how the company uses agents to automate dealer support inquiries and identify security vulnerabilities. G. V. Prakash Kumar of Estée Lauder described deploying agents to replicate makeup artist expertise for consumers by integrating real-time inventory and trend data.

The speakers agreed that organizational accountability rests with the enterprise rather than the model provider. They identified several barriers to widespread adoption, including the requirement for clean data foundations, the rapid pace of technological evolution, and the cultural challenges associated with change management and workforce reskilling.