Uploaded February 2026 | Updated November 2026, 3 hours ago
Right after Davos, I caught up with Shannon Duffy, CMO at @Okta, to discuss an important topic. If AI agents now have access, autonomy, and authority, why aren’t we managing their identities as rigorously as we manage employees? #OktaPartner
Fresh off conversations that we both had during #WEF26, we focused on the identity layer behind AI agents, who provisions them, what they can touch, and who can turn them off when things go wrong.
Watch the video to learn why every AI agent must have an identity with ownership and controls, how unmanaged agent access creates hidden risk, and what it really takes to build trust in a multi-agent world.
Right after Davos, I caught up with Shannon Duffy, CMO at @Okta, to discuss an important topic. If AI agents now have access, autonomy, and authority, why aren’t we managing their identities as rigorously as we manage employees? #OktaPartner
Fresh off conversations that we both had during #WEF26, we focused on the identity layer behind AI agents, who provisions them, what they can touch, and who can turn them off when things go wrong.
Watch the video to learn why every AI agent must have an identity with ownership and controls, how unmanaged agent access creates hidden risk, and what it really takes to build trust in a multi-agent world.










