Nvidia’s investments in photonics highlight a growing shift within AI infrastructure. The question is no longer simply how to build faster processors, but how to connect, package and integrate increasingly complex systems. The next bottleneck may emerge not inside the chip, but in everything surrounding it.
Nvidia is pouring billions into photonics as the future of artificial intelligence shifts beyond the GPU. Yet a growing debate suggests the next bottleneck may not lie in silicon photonics itself, but in the packaging, assembly and integration systems needed to connect AI infrastructure at unprecedented scale.
Artificial intelligence is no longer a weightless software story. As capital markets confront energy constraints, grid bottlenecks and front-loaded infrastructure costs, AI is being repriced as heavy industry — where scarcity, not scalability, determines power and valuation.
Last Friday at Nieuwspoort, former ASML CEO Peter Wennink presented his report, “The Netherlands: From Delay to Action”, outlining a roadmap for the country’s economic future. Central to his vision is a shift toward a technologically advanced and highly productive society, supported by top-tier education and resilient economic structures. Wennink emphasizes that the Netherlands cannot rely on incremental policy adjustments alone; achieving long-term prosperity requires decisive action to secure technological leadership and strengthen national earning capacity.
