SailPoint's Mark McClain: AI agent identities need unified governance
SailPoint CEO and founder Mark McClain told Hong Kong and Taiwan media during a late-September visit that enterprises are moving toward a "hybrid workforce" of humans and AI agents, and that the core challenge is building a unified control plane that covers both human and non-human identity risk. He said many companies hold large numbers of ungoverned non-human identities, including API tokens, service accounts, software bots and autonomous agents, sometimes outnumbering human identities by 100 times, and cited a recent Hugging Face incident involving AI as an example of missing detection, governance and blocking controls. SailPoint said it shipped related security offerings in May and October and acquired Israeli startup Entro in June, whose technology it says can discover roughly 1,200 distinct non-human identity types; McClain said the company manages hundreds of millions of identities, retains 97% of customers and expects Asia to contribute about 15% of annual recurring revenue.
As enterprises deploy AI agents at scale, identity governance is expanding from human employees to machine identities — a shift that will shape whether agentic AI can be rolled out safely.