🎧 Spotify
Alex Bovee, CEO and Co-Founder of C1.ai and former identity industry leader, takes the hot seat — tackling rapid-fire questions on Agentic AI, Non-Human Identities, identity security, and the future of the digital workforce. Listen to how Alex responded to the questions below:
Have We Built Identity Security for a World That No Longer Exists?
Alex agrees — explaining that the surface area of modern businesses has shifted far beyond humans accessing SaaS applications. Today it includes agentic identities operating at machine speed, deploying workloads and accessing data in ways that demand a fundamental shift from orchestration to runtime security.
Are CISOs Ready for the Fleet of Autonomous Digital Employees?
Alex doesn’t mince words: nobody is. Speaking from a client site, he describes the reality he’s seeing firsthand — credentials on endpoints, vibe-coded MCPs, and agents deployed across organisations with little oversight. He captures the tension well: board-level pressure to adopt AI has never been greater, but the sprawl of what now needs securing has fundamentally exploded. In his words, it feels a little like racing towards a cliff.
Is IAM for AI More Important Than AI for IAM?
Alex believes IAM for AI is the bigger imperative. While AI for IAM helps automate manual identity processes, the more significant architectural shift is ensuring agents have a centre of gravity — a control point through which they can be authenticated, authorised, and monitored in real time. For Alex, identity is the linchpin of that.
Are AI Agents Being Deployed Without Security Guardrails in Production?
One word: 100%. No elaboration needed — and the fact that he’s saying it from a live client engagement makes the point land harder.
Does Agentic AI Security Equal NHI Security?
Not necessarily. Alex points out that NHI means different things to different people — service accounts for some, API keys and tokens for others. Agentic identity, in his view, is its own category of problems: some aspects resemble delegated human authorisation, others look more like service accounts. The boundaries blur, and that ambiguity is precisely what makes it difficult to govern.
Do AI Agents Behave More Like Employees or Software?
Both — and that’s exactly what makes them hard to manage. Alex frames it clearly: humans are non-deterministic but trustable, software is deterministic and trustable, agents are non-deterministic and non-trustable. Without a moral compass or reputation at stake, agents will simply maximise toward their goal by any means available. He predicts a resurgence of insider threat detection technologies — tools like UEBA — because the untrustability of agents makes that category more relevant than ever.
Will AI Agents Eventually Need to Have Proper Background Checks Like Humans?
No — Alex pushes back on the framing. Applying traditional HR concepts like background checks to agents doesn’t quite make sense to him. What does make sense is understanding the mission intent and scoped goals of an agent before authorising its actions. That, he argues, is the digital equivalent.
Should AI Agents Be Allowed to Create or Provision Other Identities?
Yes — and Alex notes they kind of already do. Many agent harnesses already support sub-agents and federated work. He sees it as an inevitability driven by the need for parallel processing and multitasking. The key isn’t preventing it — it’s ensuring the same control points apply.
Will the AI Agent Digital Workforce Lead to Mass Unemployment?
Alex is firmly not an AI doomer. He draws on the Industrial Revolution as a reference point — a transition that took society from 90% agrarian to 90% urban and created entirely new categories of work in the process. He expects the same here: new roles like agent flight controllers and agent monitors that we can’t fully contextualise yet. The shape of work will change, but mass unemployment isn’t the outcome he foresees.
Would You Trust an AI Agent to Make a Decision That Could Impact Your Own Security?
No. Simple, unequivocal, and — given everything he’s just said — entirely unsurprising.
NHI Foundation Level Training
If you work with AI agents, cloud services, APIs, or CI/CD pipelines, securing Non-Human Identities is essential.
Take the most comprehensive and only accredited NHI training program in the industry and learn how to manage and secure NHIs, including AI agents.
Enrol here – https://training.nhimg.org/