As summer vacation season gets underway, cybersecurity officials are warning of cybercriminals weaponizing generative AI to create look-a-like booking websites and cloned brand messages.

Kayne McGladrey, IEEE Senior Member, framed the organizational failure in business terms. "Anything that seems to have a cybersecurity flavor is generally put into the cybersecurity risk category, which is a complete fiction. They should be focused on business risks, because if it doesn't affect the business, like a financial loss, then nobody's going to pay attention to it, and they will not budget it appropriately, nor will they adequately put in controls to prevent it," McGladrey told VentureBeat. That logic explains why session governance, token lifecycle management, and cross-domain identity correlation fall into a gap between IAM and SecOps. Nobody owns it because nobody has framed it as a business loss.

As generative and agentic AI systems transition from experimentation to enterprise-scale deployment, executive leaders are redefining how security, governance and innovation intersect. This panel brings together leaders and senior practitioners for a candid discussion on adopting AI security best practices across complex organizations.

Before managing risk, a new leader must establish three basic documents and secure agreements with the other business leaders to write down their decisions. These documents help connect technology directly to revenue streams and operational continuity, and make cybersecurity a business function grounded in reality. Sample document templates are included with this article for readers to adapt for their own needs.

Kayne McGladrey, senior member of the Institute of Electrical and Electronics Engineers (IEEE), echoed the concern about VS Code extensions running with full trust, "which means that they get access to the developer's filesystem, credentials, cloud keys, SSH keys, and environment variables."

Seven surfaces. One group confirmed across at least three of them, with open-sourced tooling enabling copycats across the rest. Kayne McGladrey, IEEE Senior Member, told VentureBeat that organizations are "defaulting to cloning human user profiles for agents, and permission sprawl starts on day one." The compliance frameworks enterprises rely on were written for humans. Agent identities do not appear in any control catalog McGladrey has encountered.

Kayne McGladrey is a globally recognized cybersecurity expert, known for his work in AI governance, risk management, and enterprise security strategy. A senior IEEE member and consistently ranked among the top thought leaders by Thinkers360, Kayne has contributed extensively to the industry through media appearances, webinars, and advisory roles across multiple regions.

Series C startups (and earlier), and firms in less-regulated industries (like a car dealership trying to deal with GLBA), are the best candidates for a vCISO model. These organizations often lack the budget for a six-figure executive but still have compliance mandates that require a designated security leader. A vCISO represents the necessary strategic oversight and audit readiness without the overhead of a full-time salary and benefits.

Independent practitioners confirmed the pattern across RSAC 2026. Kayne McGladrey, an IEEE senior member, told VentureBeat that organizations default to cloning human user profiles for agents, and permission sprawl starts on day one.

Kayne McGladrey, an IEEE senior member who advises enterprises on identity risk, described the same dynamic independently in an interview with VentureBeat. Enterprises are cloning human permission sets onto agentic systems, McGladrey said. The agent does whatever it needs to do to get its job done, and sometimes that means using far more permissions than a human would.

Kayne McGladrey, an IEEE senior member, told VentureBeat that organizations are defaulting to cloning human user profiles for agents, and permission sprawl starts on day one. Carter Rees, VP of AI at Reputation, identified the structural reason. "A significant vulnerability in enterprise AI is broken access control, where the flat authorization plane of an LLM fails to respect user permissions," Rees told VentureBeat.

McGladrey's practitioner experience confirms the gap. The Cloud Security Alliance published an NIST AI RMF Agentic Profile in April 2026, proposing autonomy-tier classification and runtime behavioral metrics. But SOC 2, ISO 27001, and PCI DSS have not operationalized agent identities. The compliance frameworks McGladrey works with inside enterprises were written for humans. Agent identities do not appear in any control catalog he has encountered. The gap is a lagging indicator; the risk is not.

The list of possible impact scenarios is extensive. Instead of trying to identify them all, McGladrey advises identifying the most likely and most representative types of incidents and then focusing on how such incidents could impact the business. From there, leaders must determine what impacts would be intolerable based on the organization’s risk tolerance.

"This is something that can be done because we don't have a federal privacy act in the United States, whereas in other countries, this would be completely unacceptable as well as considered to be culturally unacceptable," McGladrey said.

So it was Tom Cruise waving his hands around to use a computer, but it showed a world where people got arrested for crimes they hadn't committed yet based on data that could be flawed or biased. And it turns out that movie was a warning.

Kayne McGladrey raises another concern: the transfer of risk. AI vendors may provide the tools, but providers often carry the legal and financial consequences when things go wrong. In already stretched rural systems, that imbalance could have serious implications. There is also the issue of data. Many AI models are trained on urban populations, which may not reflect the realities of rural patients. That increases the risk of misdiagnosis or ineffective recommendations, particularly in communities with different health profiles.

Kayne McGladrey, an IEEE Senior Member who advises enterprises on identity risk, made the same diagnosis in an exclusive interview with VentureBeat. "It uses far more permissions than it should have, more than a human would, because of the speed of scale and intent."

DORA compliance requires both proper documentation and comprehensive data generation. The gap between policy and practice can be bridged by rigorous, automated evidence collection alongside documented ICT risk management frameworks. But as agentic AI continues to redefine modern operations, the definition of sufficient evidence must similarly modernize. Organizations that adopt JIT access, unified logging, and agent-specific telemetry today will not only survive the next NCA audit, but will also achieve longstanding operational resilience.

"You can reduce risk," said cybersecurity expert Kayne McGladrey. "But nobody out there can be perfect. It's an unattainable goal." McGladrey said he tends to think of cybersecurity in terms of risk; sometimes the risk is increased, and some things decrease risk.

This guide is for compliance officers, technical leads, CISOs, and their legal advisors preparing for increased regulatory scrutiny. Organizations must prepare for potential reviews of their risk management systems, data governance, and cybersecurity measures. Failure to provide adequate documentation may result in significant administrative fines, making the preparation of sufficient evidence a top priority for legal and technical teams alike.