Hitachi executive calls for AI standards, regulation and human oversight

At TechEx in San Jose, Hitachi’s Simon Ninan told CAIO Connect host Sanjay Puri that AI’s next phase depends on regulation, safety and human-centered deployment. He warned that physical AI and agentic systems need standards to scale without creating workforce gaps or safety risks. Why it matters: - AI adoption is moving from software into physical systems where failures can affect workers, infrastructure and public safety. - Ninan said the long-term winners will be companies that balance innovation with responsible governance and keep human expertise in the loop. - The debate now reaches beyond productivity gains to questions about regulation, accountability and workforce design. What happened: - Simon Ninan of Hitachi joined CAIO Connect Podcast host Sanjay Puri at the TechEx conference at the San Jose Convention Center. - The conversation covered artificial intelligence, workforce disruption, physical AI and industry-wide governance. - Ninan told the audience of chief AI officers, enterprise leaders and technology professionals that AI’s rapid advancement is inevitable. - He said the bigger challenge is making that progress safe, regulated and human-centered. The details: - Ninan described his path from early interest in AI and inspiration from his father’s work at the Indian Space Research Organisation to leadership roles at Deloitte and Hitachi. - He said Hitachi, a 115-year-old industrial and technology conglomerate, sits at the intersection of operational technology and information technology. - Ninan linked Hitachi’s longevity to “responsible innovation,” which he said prioritizes people, institutions and sustainable growth. - He called workforce transformation “the existential question of our times.” - Ninan said AI may replace some jobs, but companies should not eliminate the entry-level and middle-management layers that develop future leaders. - He warned that losing those roles without a plan could create a future skills vacuum. - He said organizations will eventually need to preserve human expertise so technology serves people, not the reverse. - The discussion highlighted physical AI, including autonomous machines, industrial systems and other real-world applications. - Ninan said physical AI needs far more predictable and reliable performance than generative AI because safety is directly at stake. - He said industries using autonomous vehicles, heavy machinery, power infrastructure and robotics need near-perfect reliability before scaling. - Ninan said many current physical AI applications are still experimental. - He said the next decade will focus on moving those systems from controlled demonstrations into complex real-world environments. - Ninan said companies must take responsibility for the outcomes of the technologies they deploy. - He called for industry leaders and governments to work together on standards and regulatory frameworks for AI, especially physical AI. - He said the lack of standards leads to fragmentation, higher costs and slower adoption. - He argued that regulation is a foundation for trust and large-scale deployment, not a barrier to innovation. - Ninan said companies should use partners for specialized AI capabilities and focus on their own strengths. - He pointed to Hitachi’s partnerships with OpenAI, Anthropic, Microsoft, Google and Nvidia. - He said Hitachi’s differentiator is deep operational and domain expertise. Between the lines: - Ninan’s comments reflect a broader shift in enterprise AI from experimentation to governance and operational discipline. - His emphasis on standards suggests that physical AI may face the same scaling problems that earlier industrial technologies encountered without common rules. - The workforce message was also a warning: efficiency gains can backfire if companies do not rebuild talent pipelines. - The partnership strategy signals that enterprise AI may increasingly reward firms that combine outside model capabilities with in-house domain knowledge. What’s next: - Ninan said the next decade will test whether physical AI can move safely from pilot projects into widespread deployment. - Industry efforts around standards and regulation are likely to shape how quickly autonomous systems scale. - Hitachi and other large industrial players are expected to keep expanding AI partnerships while refining human-centered deployment models. The bottom line: - Ninan’s message was clear: AI will advance, but sustainable adoption will depend on standards, safety and keeping humans central to the system.