Parm Sandhu, Group Vice President of Enterprise 5G Products and Services at NTT DATA, said that while global interest in GenAI grows, the importance of physical AI cannot be ignored, and physical AI will be the next AI inflection point. Global AI adoption remains uneven, with the gap now about infrastructure and governance rather than model capability: Japan offers applied-engineering rigour and a deep industrial base for domain-tuned models like tsuzumi and edge deployments, China is compressing the cost curve on open-source models through Qwen and DeepSeek, and South Korea leads on AI-ready hardware. Across APAC, research shows a widening divide, with leading organizations operationalizing AI securely, responsibly and at scale. Sandhu said most AI projects stall for three reasons: teams treat AI as a portfolio of use cases rather than a business transformation; data and infrastructure foundations are not ready, with around one in three Chief AI Officers identifying building and integrating models in private or sovereign environments as their single biggest adoption barrier; and IT and OT operate in silos, blocking operational data physical AI needs. NTT DATA has committed US$ 1.5 billion to AI-ready data center infrastructure in India, focused on sovereign compute and secure data residency. Sandhu said GenAI amplifies how enterprises think, while physical AI transforms how they operate, moving intelligence into manufacturing, logistics, energy and mining; work with Hyster-Yale embeds physical AI into a manufacturing assembly line, cutting deployment timelines