Telecoms.com periodically invites expert third parties to share their views on the industry’s most pressing issues. In this piece Nathan Howe, Vice President of Emerging Technology at Zscaler, examines ways in which artificial intelligence will affect the telecoms industry.
Whether you believe it is the answer to all our problems or a threat to life as we know it, there can be no argument about the rising influence of AI on the world around us. And seemingly no industry, service or solution that is immune to its promise of change.
With its intertangled web of processes, technologies, players and demands, connectivity is an area ripe for AI involvement. Gartner predicts that by 2025, 60% of enterprises will be using five or more wireless technologies simultaneously. It also expects that by 2027, more than 50% of the global population will be active daily users of multiple superapps – applications that provide “end users (customers, partners or employees) with a set of core features plus access to independently created miniapps.” And these are just two examples of how the current web is only going to get denser.
But what role can AI play in cutting through all this complexity and enabling the future of connectivity? In this article we are considering AI’s impact on three distinct areas – infrastructure, services and end user consumption.
How enterprises approach their infrastructure
Over the past few years, we have already started to see a shift away from infrastructure being thought of as just a process enabler i.e., helping to run a service. Instead, today’s enterprises are increasingly coming to expect infrastructure to deliver the service, without any additional help required. In basic terms, the brief to the technology vendor is changing from “I want to do this, how can you support me” to “I want to do this, make it happen.”
Now initially, much of this work will be facilitated by pre-defined functions and service templates – things that can be chosen from a vendor’s service catalogue as it were. But over time more and more will become AI-driven – with AI being used to create, run and most crucially optimize services that have been tailored to that company based on the vendor’s experience and expertise. For customer comfort, humans will still need to be involved in the delivery of these services i.e., for oversight and maintenance, but even those that are not AI-driven will at the very least be automated so they run intelligently.
How services are delivered to users
Beyond service automation, another connectivity area where AI will have a far bigger impact is in the dynamic assessment and delivery of services. By this I mean not just being able to tell if something is working or not, but knowing how to fix it if it isn’t working, improve and get it to users in the most effective way possible (i.e., optimization).
With people never more mobile, applications themselves will also need to become more intelligent in order to keep up and organisations will look to AI to help them achieve this. The most extreme example I like to give for this would be to consider the case of someone using a Tokyo-based Augmented Reality application via 5G on a Japanese bullet train. Sitting in the train station their experience is going to be great, but as soon as that train starts moving the application is going to get slower and slower as its signal has to travel further and further back to the base. With speeds of up to 320 km/h, that distance will rack up in a flash.
This requirement for applications to move with the user as they traverse cell zones (versus their more static state today) is something Zscaler’s 5G team and broader industry has been discussing for some time now – and we expect it to become a reality sooner rather than later.
How the end user consumes services
AI’s use for making connectivity more intelligent is something that will also carry through into how the end-user consumes services – or more specifically how delivery of these services is switched between different infrastructures (i.e., 4G / 5G / Wi-Fi / Bluetooth) to ensure a seamless user experience.
Devices today already carry the ability to route packets across these various connectivity types, but they aren’t using much intelligence to flip from one to another – the process is pretty sequential i.e. if this isn’t working, try the next. With AI, devices will be able to monitor these connectivity types and determine which is the best to use at any one time, regardless of whether they are next in line.
Like the application path, the user path will become extremely dynamically driven, based on performance versus having to follow a set route through the underlying infrastructure. Rather than representing siloed pathways, the underlying infrastructure will instead function more as a mesh that the services can run on top of.
For the user, this means not having to think at all about the way they connect – and likely consuming more services as a result (to the benefit of providers).
An optimized future experience
Though much of the current public discourse relates to whether or not we will ever feel comfortable engaging with AI, AI’s real role in the future of connectivity will centre around optimization versus interactivity – delivering intelligence that is constantly iterating and improving services for the benefit of those who use them.
The end result will be infrastructure that is everywhere and optimized to deliver, services that are as quickly and successfully available to the user as possible, and end user satisfaction of being able to consume what they want, when they want and where they want. It is a compelling vision, and one I am looking forward to seeing realized.
Nathan Howe is Vice President of Emerging Technology at Zscaler, where he is in charge of developing the security strategy for 5G. As a digital transformation and telecommunications expert, he’s assisted hundreds of enterprise customers seamlessly modernize their environments, adapting to distributed workforces, cloud migrations, and standing up private 5G networks. He has +20 years experience in the security industry and previously worked for Nestle & Verizon.