Ronnie Vasishta, NVIDIA (00:21):
Yeah, one of the most exciting areas is the way operators are going to use AI to change their services, whether it be digital humans on a phone as personal assistance, or it be the connection of autonomous vehicles. Obviously video and the enablement of talk to your video, make sure that you can see and collect the right activities that are coming off video. This is an incredibly exciting area for how now generative AI is going to be deployed by operators to their businesses.

Alex Choi, AI RAN Alliance & SoftBank (01:02):
When integrated with large language model LLM, we can expect to see relatively visual improvements in areas like a customer service touchpoint. Additionally, traditional communication services such as messaging can be upgraded to a new level of customer experience through AI and large language models. So there is also high possibility of creating a specialized services such as public safety, which are difficult for existing OTT platforms to support.

Patrick Kelly, Appledore Research (01:45):
I think the Edge network is probably the last frontier for communication service providers to capitalize on monetization of the network assets and investments that they've made. And in this domain, it's perfectly reasonable to see AI inferencing being applied to the edge network for workloads and services. So this could be done via revenue share model, or you could at its most basic level, you could lease out central offices to hyperscalers or virtual specialists or integrators to increase potentially billions of dollars in revenue. Coming back to communication service providers,

Phil Cutrone, HPE (02:34):
Well, I actually think both use cases are linked but not the same. And that is on the for business customers, I think it presents an opportunity for the operators as they deploy private networks like private 5G, that enterprises or business customers are going to require AI workloads to be run. And the question is where and for their low latency. And I think the operators have a tremendous opportunity to push compute resources to the far edge to address those private networks specifically for that enterprise. Now, I also think for the consumer presents a lot of opportunity for the operator in that there are going to be generic kind of applications that we use for inferencing. Some of them can just be fun that we use as an application that we download, but some might actually be practical use cases and most likely that's where the opportunity and revenue opportunity presents itself. So then in that particular scenario, it's still moving compute resources to the far edge, so now the operator can participate as opposed to having over the top services being delivered by another company like some of the hyperscalers or other cloud-based services. So I think the operators presented a tremendous opportunity to capitalize on both business customers and even consumers.