Today, new research from Cequence Security, the leader in application, API, and agentic AI protection, and Enterprise Management Associates (EMA) found that 94% of enterprise IT and security leaders are confident their AI agents do not have more access than they need, yet only 33% actually provision agents with least-privilege access. The remaining two-thirds run on broad standing permissions that are reviewed periodically, rarely reviewed, or never reviewed at all.

That gap between confidence and practice is already showing up in production, not a theoretical risk, but as incidents enterprises are living with right now. Among the organizations surveyed:

• 65% have experienced an AI agent take an action outside its intended scope, including 29% with measurable business impact, including data exposure, financial loss, operational disruption, or reputational damage. Another 36% caught a near-miss before it caused damage.

• Only 32% can detect and contain an out-of-scope agent action within minutes through automated means; 55% need hours and manual steps to respond.
• In approximately 4% of organizations surveyed, the first sign of trouble came from a customer or outside partner, not an internal system.

The findings point to one clear story. Governance has not kept pace with the speed of agentic AI deployment, and that gap is showing up at every stage of the agent lifecycle, from how agents are provisioned, to how their actions are authorized, to how they are decommissioned once a pilot ends. Other key findings from the report include:

Enterprises Have Moved Past the Pilot Stage

The scale of deployment makes the gap more urgent. 46% of organizations report they are already scaling agentic AI across multiple departments and production workflows, and 79% are running generative and agentic AI simultaneously. Further, more than 92% report an increase in AI and bot-driven traffic targeting customer-facing applications and APIs.

Authorization is Checked at the Wrong Time, Or Not At All

That governance gap extends to how access is enforced in the moment an agent acts. Only 34% of organizations evaluate an AI agent’s authorization at the moment it attempts a specific action. The majority rely on periodic policy reviews or standing permissions set once at provisioning and never revisited, meaning an agent’s access can quietly outlive the task it was originally granted for, and keep working long after anyone signed off on it.

Abandoned Pilots Are Leaving Live Credentials Behind

Additionally, there’s an increasing risk in how enterprises manage agents that don’t make it to production. 31% of agentic AI pilots have been paused indefinitely, discontinued, or abandoned. Many were real deployments with real system access and credentials that were never cleaned up. Every abandoned pilot with live credentials is exposure nobody is actively watching.

External Connectivity Carries the Same Risk

14% of organizations allow AI agents to connect to outside tools and data sources via the Model Context Protocol (MCP) without restriction. Among the majority who do limit those connections to an approved list, fewer than half, just 49%, have a dedicated team actively maintaining and auditing that list on a regular basis.

Christopher M. Steffen, CISSP, CISA, VP of Research at EMA, said: “This research shows enterprises have moved well past experimentation with agentic AI right into production, and governance has not kept pace with that shift. The gap isn’t a lack of awareness; most organizations have policies in place and express real confidence in them. The gap is between what’s written down and what’s enforced when an agent takes an action nobody approved. That disconnect shows up most clearly in how organizations authorize agent actions and monitor them once they’re live, and it’s the reason incidents are happening at a rate the industry hasn’t fully reckoned with.”

Shreyans Mehta, Co-founder and CTO at Cequence, said: “The number that jumped out to me is the 92% being confident in their governance frameworks. Confidence like that is a trap; it’s exactly why organizations stop looking for problems, stop investing in monitoring, and let authorization checks lapse until an incident forces the conversation. This is the exact blind spot Cequence is built to close, giving security teams real-time visibility into what AI agents are actually doing and enforcing authorization at the moment an agent acts, not after the fact.”

Elaborate on the API features of the platform and their efficiency to accommodate existing software resources of enterprises

While designing our product we always have had an API-first approach. The way we have architected and designed our product is to keep it developed friendly and extensible and API is a major part of it. Every functionality of our product is supported by APIs. We have more than 680 APIs currently. These APIs can be used by customers to extend the functionality, automate a lot of tasks, create custom flows, etc.

Explain the dashboard and reporting features you offer in the platform

In Helical Insight, we offer a very simple, no-code, self-service, and drag-drop interface allowing users to do their ad-hoc analysis. We wish to empower users to not depend on technical people and create their analyses. People can drag, drop required fields, add filters, select any of the 30+ visualizations, and customize them. With our simple interface, people can create reports, dashboards, infographics, map-based analytics, etc.

Our range of visualization options includes table, crosstab, axis charts, non-axis charts, maps, advanced statistical and scientific charts (bubble, dendrogram, treemap), polar, radar, etc.

How efficiently can they use for or impact business decisions? Give a scenario, if possible

Nowadays every company is using so many data sources and generating a lot of data. These data points are not having any value if the organization is not able to derive insights from this data and that is where we come in. We help companies to derive meaningful insights from this data which can in turn help them to decrease their costs, improve their bottom line and revenues, assign their resources in places/products with higher margins, and much more.

Nikhilesh Tiwari, Director – Helical Insight

Use case: We have a client who is in the energy & utility domain and they have built IoT devices, which are deployed on the poles and transformers. From these devices, they collect a lot of data about the transmission and distribution, current, voltage, temperature, pressure, wind speed, etc. With our BI product Helical Insight we are helping them to identify the input and out, various kinds of real-time analysis, identify the pilferage and thefts, trends of usage, generate alerts as well as predict the chances of a transformer going down.

Tell us about the client base you have in the Middle East and plans for the upcoming years

We have customers all across the world, including Australia, Indonesia, the Middle East, Israel, India, the US, South America, etc. though most of th customers are from US and India.

As of now, we are having around five clients from the Middle East that focus on education, retail, and manufacturing domains. With our 5.0 release, we expect exponential growth in the customer base. We want to have a much larger footprint in the Middle East and for the same, we have started sponsoring and exhibiting various events in the Middle East. We are also in advanced talks with some companies in the Middle East and exploring partner programs with them.