Ankabut is evolving beyond its traditional role as a national research and education network to become a key enabler of the UAE’s AI ambitions. Adli Dehelia, Executive Director, Service Development, Ankabut, discusses the organisation’s transformation and its vision for AI-native education.
How has the organisation evolved from a national research and education network into an enabler of the UAE’s AI ambitions?
You can think of us as Ankabut 2.0. That’s how we like to describe ourselves.
Since the new management came in, our targets and objectives have changed. We have evolved from an organisation built around the National Research and Education Network (NREN) and the underlying infrastructure into one that covers the entire technology stack, from networking to user experiences and journeys.
We are approaching this in two ways. First, when it comes to systems of record, we are not trying to reinvent the wheel. We have partnerships with external companies that allow us to provide those solutions.
Instead, we are positioning ourselves as an orchestrator. We want to use AI to bring different services together and create better experiences for students, faculty members and employees.
One of the challenges in education is that the sector has not transformed as quickly as some other industries. Academia traditionally takes longer to embrace change.
What we are trying to do at Ankabut is connect the dots between teaching, learning and the smart campus concept.
For example, when a student opens an application, they should not just see their grades or academic performance. They should have a complete 360-degree view of their campus experience.
That could include wayfinding, access to laboratories, catering facilities, meeting rooms and other campus services. Everything should work together in an automated smart campus environment, alongside teaching and learning.
We are also bringing AI into this environment through practical use cases. We have a data platform that allows us to generate insights, understand student behaviour and identify situations where action may be needed.
The idea is to create a complete framework that educational institutions, universities and even K-12 and continuing education providers can benefit from.
But the important point is that we want to use AI where it genuinely adds value, rather than simply adding AI features for the sake of it.
We call this AI in action. It is about building AI-native solutions rather than taking traditional systems and adding a few AI capabilities.
That’s the new vision for Ankabut.
Some of the products we develop will be taken to global markets, while others will be delivered as services for the local market.
We are also investing in infrastructure, including AI labs and high-performance computing.
For example, researchers at smaller universities may need GPU capacity or high-performance computing infrastructure for specific projects. Rather than investing in their own expensive infrastructure, they can access our resources for the period they need them.
They can also benefit from our high-bandwidth network infrastructure.
Do you think the UAE has sufficient infrastructure capacity to support the AI-native education environment that Ankabut is developing?
I think the UAE has been one of the leaders in this area. The country started investing in digital infrastructure a long time ago.
Today, we have the major hyperscalers operating in the market, alongside organisations like Ankabut that provide private cloud infrastructure for specific sectors.
There is already a strong infrastructure foundation.
Of course, as demand for AI and GPU capacity continues to increase, infrastructure investment will also need to keep pace.
But overall, I believe the UAE has the right infrastructure and investment environment. We simply need to continue building on that foundation.
What does AI-native education mean to you and how is it different from simply adding AI features to existing education platforms?
Traditionally, educational institutions have relied on systems of record, such as student information systems.
What we are seeing now is that vendors are adding AI features to these existing platforms.
But AI needs data from multiple systems to deliver meaningful results.
For example, if you want a complete 360-degree view of a student, you need to bring together information from different sources.
That’s why our approach is not to focus on individual applications. We focus on orchestration, connecting systems and using the information they generate to provide meaningful insights.
A student needs to understand more than just their grades. They need information about attendance, courses, academic progress and areas where they may be struggling.
Let me give you an example.
Traditionally, identifying a student who is at risk involves generating reports and reviewing academic performance to determine who may need additional support.
In our environment, AI can identify that a student is struggling with a particular chapter or subject.
That information can then be shared with the student’s academic adviser, teacher and the student themselves.
But we don’t stop there.
The system can recommend a specific action. For example, it might suggest that the student revisit Chapter Eight and use an AI tutor for additional self-learning.
So, we are looking at the complete process, from identifying a problem to recommending an action and supporting the student.
That’s what we mean by AI in action.
It is not about using AI to perform one isolated task. It is about making sure that different systems and services work together in a coordinated way to improve the overall learning experience.
With the growing adoption of agentic AI, do you foresee a future where AI agents could replace human teachers?
I don’t think we will reach a point where human teachers are no longer needed.
What AI can do is reduce the administrative workload that teachers deal with every day.
Teachers spend a considerable amount of time on administrative activities, reviewing student performance and identifying where individual students need support.
AI can take over some of those tasks and help teachers work more effectively.
And it is not just teachers who will benefit. Academic leaders, registrars and other university employees can use AI to improve their productivity.
It gives them more time to focus on other important responsibilities.
Of course, AI will affect jobs. We often hear concerns about people losing their jobs because of AI.
But I think the bigger issue is skills.
People will need to develop new capabilities and learn how to work with AI.
Those who choose not to improve their skills or adapt to these changes risk being left behind.
How is Ankabut’s relationship with cloud providers and hyperscalers evolving as you expand your AI infrastructure capabilities?
Regardless of how much we invest in our own infrastructure, we will still need hyperscalers. For example, we are setting up an AI lab with a certain amount of GPU capacity.
But there will be times when we need additional computing resources.
We are already working with hyperscalers and investing in orchestration tools that will allow us to use our on-premises infrastructure and hyperscaler resources together.
The objective is to make this process transparent to users.
They should not have to worry about exactly where a workload is running.
If we need additional capacity, we should be able to use hyperscaler infrastructure without disrupting the user experience.
This is how we see our infrastructure evolving into a hybrid environment, where our own resources and those of hyperscalers complement one another.
Universities and research institutions hold large amounts of sensitive data. With sovereign AI becoming an important priority, particularly in public sector education, how is Ankabut helping institutions keep their data within trusted environments?
This is an important part of what we do. We have our own infrastructure that is dedicated to serving educational institutions.
At the same time, the multicloud approach allows organisations to keep their data in one location while using computing resources or services from other environments.
The key is to ensure that only the data that genuinely needs to be transferred is moved, and that this happens under strict governance controls.
We are putting the right platforms and controls in place to support this. If certain data needs to remain within a particular institution, we make sure it stays there.
Today, it is increasingly possible to access and work with data across hyperscalers, private clouds and local infrastructure without necessarily moving that data between environments.
That is the direction we are taking. Our objective is to give educational institutions the flexibility to use advanced AI and cloud capabilities while maintaining control over their sensitive information.
