Why Classical Optimization Struggles to Keep Pace With Modern Manufacturing
May 5, 2026 | 5 minute read
Read article
Space Isn’t the Only Way To Address AI’s Energy Challenge. Quantum Can Help, Now.
April 6, 2026 | 5 minute read
Article Highlights
The Search for Computing Power Expands Beyond Earth
The world’s largest technology companies are converging on a striking idea: if AI is running out of power on Earth, expand compute beyond traditional terrestrial infrastructure. Leaders at SpaceX, Google, Nvidia, and Amazon are exploring bold approaches to next-generation AI infrastructure, including orbital data centers powered by continuous solar energy and cooled by the vacuum of space. The logic is compelling. AI systems are consuming unprecedented amounts of electricity, straining terrestrial grids, water resources, and permitting processes.
In theory, space-based infrastructure could open up new possibilities for abundant energy, cooling, and scale. At the same time, these efforts will take time to mature. Building large-scale compute infrastructure in orbit is an ambitious undertaking, with significant engineering, economic, and operational challenges still to be solved.
That is exactly why it is worth also focusing on solutions that can help in the near term, here on Earth.
Why Annealing Quantum Computing Is Now an Energy Efficient Computing Alternative
Quantum computing offers a fundamentally different approach to computation, one that can reduce the energy required to solve certain classes of problems. Unlike classical systems that often require increasing amounts of power as problem size grows, quantum systems can tackle some solution spaces far more efficiently. For optimization, materials simulation, and emerging machine learning workflows, quantum computing can deliver meaningful results with substantially less energy.
This opportunity is not long term. It’s now. D-Wave’s annealing quantum computers are already being applied to complex problems in materials development, manufacturing, and life sciences — problems that can require significant classical compute resources. In drug discovery, for example, D-Wave worked with Japan Tobacco’s pharmaceutical division, now part of Shionogi, on a quantum AI proof of concept designed to improve the training of generative models for novel molecular design. That is an important example of how quantum and AI can work together on real-world problems today.
On the science side, the potential efficiency gains of quantum computing are becoming hard to ignore. In a published magnetic materials simulation, D-Wave showed that a problem solved on our Advantage2™ quantum computer in minutes would have taken a classical supercomputer nearly one million years and more than the world’s annual electricity consumption to solve. Whether the use case is scientific discovery, optimization, or AI-adjacent workloads, that kind of result points to a larger truth: better computation can also mean more efficient computation.
Can Quantum Computing Reduce Energy Use Before Space-Based AI Data Centers Arrive?
The real question is not only where we should put compute. It’s also how much compute we actually need, and how we intelligently use it.
Space-based data centers may absolutely become part of the future of AI. But while that strategy evolves, annealing quantum computing offers a practical path to explore right now, helping organizations improve performance, efficiency, and energy use in the near term. We can pursue bold long-term infrastructure ideas while also adopting new computational approaches that can make a difference today.
Elon Musk, Sundar Pichai, Jeff Bezos, Sam Altman — big bets on future AI infrastructure matter. But so do practical approaches that can improve efficiency today. Quantum computing should be part of that conversation.
A version of this article originally appeared on LinkedIn.
More Articles
