IBM warns of telco AI ‘haves and have nots’

The big message coming from IBM’s stand at MWC 2024 was the recent collaboration between it and the GSMA designed to support the adoption and skills of generative AI in the telecom industry.

It’s called the GSMA Advance’s AI Training program and the GSMA Foundry Generative AI program, and it is supposed to ‘prepare telco leaders for the AI-era and bridge skills gaps in the telecoms industry,’ with the support of watsonx, IBM’s AI and data platform.

It will take place at IBM offices in five locations around the world in 2024, including Dubai, London, Mexico, New York and Seoul, and an online training program will be available in multiple languages.

Stephen Rose, IBM’s GM, Global Industries, fleshed out some of the thinking behind the initiative from the MWC show floor in Barcelona today.  

“What we were concerned with was that it was evident that there were going to be a few [operators] that were able to charge ahead with experimenting with AI, because they’ve got the money and the resources. And then very quickly, there's potential for establishing an AI gap between the haves and the have nots. So what we looked at was how are you going to solve that problem? We don't want to encourage an adoption gap. We want to accelerate the exact opposite of that. So we thought about what were the types of problems that they've got – the first is just making sure that you can help executives at second and third tier operators, the wider ecosystem, and emerging markets [find out] what’s the potential in generative AI.

“So we've established five training venues, IBM innovation centers, around the world, and we educate around what is AI, what's the difference between generative and traditional, what are the use cases for it? And as an executive, maybe what are the strategic considerations in terms of leadership, governance, how do you establish trust, what are the technical economic cases for AI, and what are the technology architecture choices that you need to think about as well?”

IBM isn’t alone in seeing generative AI as a potential ray of sunshine for telecoms firms looking for new revenue streams - Nvidia’s global head of business development for telco Chris Penrose today told us that it could finally help them recoup the costs of deploying 5G. When asked if generative AI in some form or another could open up new areas of monetisation for cash-strapped operators, Rose added:

“What we've been doing is taking access technology and saying 3G, 4G 5G – there is a benefit to each of those Gs, but the differential between 4G and 5G hasn't been sufficient enough from an access standpoint alone. And then the second thought was that cloud is going to solve for that for us – because it's going to enable us to bring in a bunch of ecosystem players. But really, it ended up becoming an efficiency discussion – should I take all of my current operations and put it into a centralized model, or a shared model, and then an ‘as a service’ model, and suddenly that that would save me money.

“But it really doesn't make me money. Maybe it gives me some ambidexterity but not a lot, really. So what's been missing is the ability to deploy different workloads wherever you want and be more creative around that - we can see hybrid cloud mesh coming in as an example – but AI really is the differentiator, because it enables levels of creativity that we've never seen before.”