What if you could deploy AI solutions on your factory floor in weeks instead of years? What if private 5G networks could deliver the real-time connectivity your manufacturing operations desperately need? In this interview from NTT Upgrade 2026, these possibilities become reality.
Paul Bloudoff, senior director of Edge AI at NTT Data, reveals how manufacturers are breaking free from traditional constraints and achieving unprecedented operational efficiency. We try to focus on solutions being deployed right now, not what may or may not happen in the future.
The conversation gets into conventional thinking about edge computing. While many define “the edge” as anything outside major data centers, NTT Data takes a different approach that focuses on getting compute resources as close as possible to production environments. The implications for manufacturing are profound, Bloudoff says.
What you’ll discover in this video
Bloudoff talks the “symbiotic relationship” between three converging technologies with us: edge AI, private 5G, and physical AI. You’ll learn why this combination is transforming how manufacturers approach everything from predictive maintenance to task verification on assembly lines.
One concrete example that stands out for us: modern physical AI models can be trained with just 40 hours of video instead of requiring 50,000 images over nine months. This acceleration changes everything about deployment timelines and ROI calculations. But there’s a catch, as always. Bloudoff explains exactly what manufacturers need to know before investing.
The discussion includes insights about pricing, which has historically been a barrier to private 5G adoption. Has the situation improved? What multiple use cases typically justify deployment? And why do safety applications often provide the initial business case?
Real-world examples you can learn from
Bloudoff shares specific examples from companies like Cargill, which has deployed private 5G across its manufacturing footprint. You’ll hear about robots replacing manual inspection tasks, digital twins improving production planning, and how the right connectivity infrastructure enables use cases that weren’t previously possible.
The conversation also addresses a critical strategic question facing manufacturers: should you wait for 6G or invest in 5G now? Bloudoff’s answer might surprise you, especially given ongoing development in 5G standards like release 17 and 18.
Key topics
- How NTT Data defines edge computing differently from telecom providers
- The full-stack approach from sensors to cloud infrastructure
- Why private 5G provides superior performance for time series data
- Physical AI applications that deliver value in weeks, not years
- Calculating ROI when safety and uptime are priorities
- Partnership ecosystems with Nvidia, Ericsson, Nokia, and specialized ISVs
- The future of automation and keeping humans in the loop
- Anomaly detection for both operational efficiency and physical security
Whether you’re evaluating private 5G for your facilities, exploring edge AI solutions, or trying to understand how physical AI differs from traditional machine vision, this conversation provides the practical insights you need. We discuss the technology and have Bloudoff explain how to measure value, justify investment, and avoid common pitfalls.
Press play to discover how leading manufacturers are transforming their operations with edge AI and private 5G, and what you need to know before following their path.