Nvidia is as hot as ever and it's not just
Verizon and
T-Mobile that embrace it;
AT&T wants a piece of the action, too.That's why Ma Bell announces a Cisco collaboration that's aimed at enhancing the evolution of AI-driven IoT. That's the combination of three technologies: Internet of Things (IoT), artificial intelligence (AI), and connectivity (such as
AT&T's networks). The goal is to turn data from connected devices into automated insights and actions in real time.
The combination
AT&T is combining its dedicated IoT network core with the Cisco Mobility Services Platform to improve how connected devices send and process data. The platform is designed to handle data locally when possible, which helps improve speed, performance and security instead of sending everything to distant cloud servers.At the same time, the Cisco AI Grid (built together with Nvidia) is meant to bring AI processing closer to where the data is actually created. This process, known as AI inference, means analyzing data with trained AI models to produce results. Instead of sending device data far away to be processed, the system can analyze it much closer to the source. This should reduce delays.
The system is designed for developers and businesses that want to use AI in real-world environments, such as factories, transportation systems or smart infrastructure. It creates a secure, end-to-end pathway where data can travel from edge devices like sensors or cameras, through the
AT&T network and into the systems accelerated by Nvidia hardware.According to Chris Penrose (Global VP, Business Development, Telco at Nvidia), distributed computing is "the next frontier" for AI infrastructure and telecommunications networks "sit at the heart of that buildout".Cisco's AI capabilities, powered by the Nvidia RTX Pro 6000 Blackwell Server Edition GPUs, play a crucial role in enhancing
AT&T's network in terms of intelligent automation. These are the same GPUs that
T-Mobile is testing, as we told you earlier. The idea is that these units should be built directly into network infrastructure so the system can handle both wireless traffic and AI computing tasks at the same time. If the tests succeed, future 5G and eventually 6G networks could rely on GPU-accelerated infrastructure like Nvidia Blackwell to handle both connectivity and AI processing at scale.