Kazu Gomi is a prominent figure in the technology industry, based in Silicon Valley. As the president and CEO of NTT Research, a division of the Japanese telecommunications giant NTT, Gomi oversees a significant portion of the R&D budget for fundamental research in Silicon Valley.
Gomi is heavily investing in artificial intelligence (AI) for enterprise applications to capitalize on the AI boom. Recently, he announced a new research initiative focusing on the physics of AI and introduced a chip design for an AI inference chip optimized for processing 4K video at a faster rate. This initiative follows previous research projects aimed at improving AI capabilities and energy efficiency in data centers.
In a conversation with Gomi, the discussion touched on the significance of physical AI in 2025, with Nvidia leading efforts in generating synthetic data to expedite the development of self-driving cars and humanoid robotics. Gomi’s team is exploring photonic computing as a means to enhance the energy efficiency of AI computing processes.
NTT, with over 330,000 employees and $97 billion in annual revenue, established NTT Research in Silicon Valley six years ago with an annual R&D budget of $3.6 billion.
The interview with Gomi highlighted the emphasis on AI in current research endeavors, particularly in understanding the inner workings of AI systems to address issues like bias. The Physics of AI research group aims to delve into the workings of AI systems to identify and rectify biases that may arise.
The discussion also touched on the development of an AI inference chip by NTT Research, designed specifically for video and image processing tasks. Unlike Nvidia’s general-purpose AI chip, NTT’s chip is tailored for specific applications like object detection in high-resolution video content.
Gomi shared insights on the potential of photonics technology in revolutionizing AI infrastructure by addressing bottlenecks in data processing and memory communication. By leveraging photonics technology, NTT aims to enhance the speed and energy efficiency of AI computations, paving the way for advancements in AI capabilities.
The conversation also covered NTT’s progress in creating an autonomous closed-loop intervention system for heart failure patients, powered by a bio-digital twin that can predict and administer precise treatments without human intervention.
Overall, NTT Research’s focus on AI, photonics technology, and innovative chip designs underscores the company’s commitment to driving advancements in AI research and development, with a keen eye on energy efficiency and performance optimization.
