As agentic AI shifts from answering questions to taking action, the hard question is no longer what agents can do...It's what they should be trusted to do on their own. Harsh Archarya, SVP of Digital Tech Operations & Infrastructure Platform Services at Dell Technologies, oversees the platforms and operations where that question gets answered every day. He will share how Dell is deciding where agents automate, where they assist, and where people stay firmly in charge, and how to build workflows that capture machine speed without giving up human judgment. 

• Mapping which decisions agents should automate, which they should augment, and which should stay entirely with people 
• Structuring review, override, and escalation paths so humans stay in control as agents take on more work 
• Setting guardrails, thresholds, and policies that keep agents aligned with brand standards, security requirements, and customer trust 
• Building the infrastructure and operational foundations that make agent behavior observable, reliable, and auditable 
• Training teams to trust agent recommendations without rubber-stamping them