We’re now a quarter of the way through the 21st century. Just 25 years ago, we feared the Year 2000 (Y2K) (or Millennium) bug would cause major computer glitches and bring everything down. Just 10 years ago — when we turned the calendar to 2015, the year Marty McFly first witnessed the hoverboard — we lamented how few of Back to The Future 2’s gadgets and tech innovations were realized.
That’s the thing about predictions: some manifest, others do not.
The most transformative tech advancements though are often those that are unexpected and invisible to the naked eye. I predict in 2025 we will see a significant evolution of business models, processes and under-the radar innovations that may not grab the headlines like Artificial Intelligence (AI) or Y2K.
Bring your own AI is here to stay
There is no doubt AI will continue to be in the news in the coming year. We have seen the excitement and hype curve for AI (particularly GenAI) throughout 2023 and 2024 and I expect it to continue well into 2025. What the future holds is how we use AI to our benefit, including how it improves our business strategies. While communication service providers (CSPs) are embracing AI, many are concerned about being locked into vendor-specific models that they cannot extend or fully understand.
This is a fear many organisations hold, not just CSPs, as a restrictive model-based approach to AI can stifle innovation by locking in costs and being overly reliant on a single vendor pushing its product forward. CSPs are looking for flexibility in managing AI use cases, including those developed internally, and in the coming year, we will start to see the democratization of AI in the telco arena.
CSPs want to bring their own AI to their processes and apply it to application vendor datasets. This will allow them to innovate to their own specific needs and on their own timeline. As AI technologies evolve and mature, an open and programmable approach to AI will give greater control to CSPs so they are not beholden to one specific vendor for their Operations Support System (OSS) tools.
To truly tap into the value of AI, an ecosystem approach where everyone takes part in the development and innovation is best. Thus, the concept of bring-your-own (BYO) AI will accelerate throughout 2025 with CSPs embracing vendors that allow third party AI to be applied against their specific requirements.
The NaaS foundation becomes more clear
Networking-as-a-Service, or NaaS, began to pick up as a trend — albeit one not as well-publicised as the AI hype train — earlier in 2024. In fact, ABI Research forecasts that more than 90% of enterprises will consume at least 25% of their network services as a usage-based consumption model by 2030.
The NaaS trend will continue apace as we enter 2025.
NaaS requires the implementation of a hierarchical Business Support Systems (BSS) and OSS model. CSPs are using this to put an end to the spaghetti integration that has hindered the sector for years. From there, APIs — both open and standard-based — must be leveraged to allow the modular OSS applications and layers to communicate consistently. We’re starting to see some standards and protocols commonly adopted across the industry. And finally, data models are the glue binding the NaaS ecosystem together, allowing CSPs to embrace declarative or intent based networking models.
The key to making NaaS work efficiently is a stable foundation — and as the proliferation of NaaS continues, the ideal architecture is starting to emerge. With a clearer picture of an ideal foundation, the use of NaaS is set to become far more common place from next year.
Speaking of OSS…
There is a growing number of legacy inventory and assurance systems reaching end of life after a 20-year lifecycle. Inventory and assurance systems developed and deployed in the early 2000s weren’t built to support dynamic services such as 5G, Edge Cloud and the Internet of Things.
This is forcing CSPs to upgrade these legacy systems so they can modernize and compete more effectively. In 2025, I predict we will see inventory and assurance become the centre pieces of CSP transformation.
Next-generation versions of these systems are cloud native and optimized for automation. They offer a stepping stone towards AI in the OSS space. Some network providers are in the process of upgrading their networks now, but I expect the coming year will bring with it a wave of greater network modernizations of these critical systems across the telco space.
The convergence of service management
Many CSPs are looking to bring Network, IT and Operations teams together to drive greater synergy and support accelerating cloud momentum. Yet for CSPs to realise the true power of this synergy, a converged services management layer enables data to be leveraged across all functions to bridge organisational divides, break down the operational silos, and reduce manual processes.
Currently, service introduction and fulfillment processes are too slow and resource intensive. This needs to change so that CSPs can speed up time-to-revenue and improve the customer experience.
In 2025, I believe we will begin to see the deployment of zero-touch service order fulfillment that dynamically discovers the network’s service capabilities, automates service activation, and provides complete order workflow visibility and control. Converging the services management layer and the data model will bring a unified approach to planning, orchestration and assurance.
We may not necessarily see the tectonic shift in 2025 that we have seen with other innovations like AI, 5G or the iPhone. And it’s safe to say many of us have given up on the idea of the hoverboard.
But the best predictions are the ones that are predictable. They allow us to plan best for the future, and they rely on the trends already taking hold to inform where we need to put our efforts. 2025 will be the year where CSPs reset their foundation, with some of these under-the-hood transformations supporting CSPs better embrace the innovations of tomorrow.
Table of Contents
