Quantum Computing and AI: Making it work

Abstract

We are experiencing a revolution in computing and are in the most exciting time since the transistor was invented. We continue to find new ways to further advance the performance of classical computing with critical innovation in semiconductor technology like lithography, new materials, architectures, packaging, photonics integration, co-optimization and more. However, an already huge and further increasing performance driver with changing workloads is artificial intelligence (AI). The compute requirements for large AI training jobs are doubling every six months demanding significant and continuous hardware and software innovation to be implemented. Therefore, we are developing new AIU processors to point the way to a future where AI computation is more efficient, less power hungry, and more capable. Besides these new AI chips a radical new compute technology is quickly advancing. For the first time in history, we are seeing a branching point in computing paradigms with the emergence of quantum processing units (QPUs). The last few years we have witnessed a strong advancement in quantum computing technologies developing the entire stack from the bottom up. We demonstrated continuous and significant performance improvements in scale, quality and speed reaching quantum processor units with more than 1000 Qubits and created a modular approach to scale beyond. Furthermore we reached the “Era of Utility” in which quantum computers run circuits beyond the reach of brute-force classical simulations. Our quantum technology roadmap is paving the way toward future error-corrected systems at the end of this decade. I will give an overview about the challenges and the recent progress to develop the future of computing with an emphasis on our roadmap for quantum computing.