It’s easy to get lost in the whirlwind of AI terminology these days – from 'Agent AI' to 'Ambient AI' and everything in between. But what does it really mean when a global giant like NTT decides to go all-in on artificial intelligence? Recently, I was looking into how major players are building their AI foundations, and a partnership between NTT Group and NVIDIA really caught my eye.
NTT, a name synonymous with telecommunications and a significant presence in Japan, is embarking on a company-wide AI initiative called 'corevo'. This isn't just a small R&D project; it's a massive undertaking involving over 5,000 researchers and 400 dedicated AI specialists. The goal? To weave AI into the fabric of their commercial products and services, focusing on real-time processing to tackle some pretty pressing societal challenges.
What struck me was their choice of platform. NTT is building corevo's cloud-based AI resource center on NVIDIA's AI platform, specifically leveraging Tensor Core GPUs like the V100. Now, I'm not usually one to geek out over hardware specs, but the numbers here are pretty staggering. These V100 GPUs, built on NVIDIA's Volta architecture, are packing 21 billion transistors and can apparently deliver the processing power of up to 100 CPUs in a single unit. That's a serious leap in speed and scalability, crucial for both training complex AI models and running them in real-time for things like call center interactions or network monitoring.
Why this particular platform? NTT sees it as the key to creating a common development environment. Masahisa Kawashima from NTT's Software Innovation Center put it quite plainly: a unified platform will allow them to effectively leverage their AI research and development efforts and speed up the creation of new value. It’s about making sure all those brilliant minds working on AI across the NTT Group can collaborate and build upon each other's successes without reinventing the wheel.
This initiative, corevo, was conceived with a clear purpose: to address issues facing Japanese society, particularly those stemming from an aging population. Think about it – a shrinking active workforce and the growing need for elder care. NTT is aiming to use AI to help.
Their approach is broken down into four key areas:
- Agent AI: This is where natural language processing shines, powering things like chatbots for customer service and even monitoring systems for the elderly.
- Ambient AI: This focuses on analyzing vast amounts of data to understand and manage things like traffic flow – making our cities smarter and more efficient.
- Heart-Touching AI: This sounds fascinating, delving into body and emotional analytics for applications in healthcare, sports, and entertainment. Imagine personalized wellness programs or more engaging experiences.
- Network AI: This is about keeping the digital infrastructure running smoothly, with capabilities for anonymous detection, predicting failures, and monitoring networks.
It’s a comprehensive vision, and by adopting NVIDIA's AI platform, NTT is essentially equipping its researchers with a powerful engine to drive these ambitious projects forward. It’s a reminder that behind the often-abstract concept of AI, there are concrete technological choices being made to solve real-world problems. And when you see a company of NTT's scale investing so heavily, it signals a significant shift in how we'll interact with technology in the years to come.
