Google Quantum AI recently announced a groundbreaking new algorithm that has the potential to transform the quantum computing landscape. This cutting-edge algorithm demonstrates an incredible advantage relative to classical computing approaches. It does the same calculations at speeds 13,000 times faster than the fastest possible classical system on the efficient Frontier supercomputer. This advance realized on 65 of Willow’s qubits could be an inflection point. It takes us one step closer to effectively harnessing the promise of quantum computing for commercial benefit.
Critics have long argued that the advantages of quantum computers over traditional machines often rely on tests involving tasks that lack real-world significance. Google’s breakthrough into quantum algorithms directly addresses these challenges. In particular, it’s able to accurately recreate complex molecular interactions, something even the most high-tech classical supercomputers have a hard time doing.
A Leap Forward in Quantum Speed
Google’s new algorithm adds a small perturbation to a qubit. In fact, this seemingly innocuous tweak can produce dramatic and far-reaching consequences on qubits that are far away. This phenomenon, wholesale, is often referred to as the butterfly effect, where seemingly minor changes produce widespread impact. In order to reach a verifiable quantum advantage, the researchers needed to utilize two quantum processors with a newly developed algorithm in parallel. This achievement represents an important step forward toward increasing the practicality of quantum computing.
In previous demonstrations, researchers realized quantum advantage using some form of random circuit sampling. They only had to get 0.1 percent of the data right to win. This new Google algorithm can only have 0.1 percent of its data be wrong. This feature greatly improves its reliability and applicability. This represents a significant step forward in the fidelity and robustness of quantum computations.
Hartmut Neven is director and founder of Google Quantum AI. He thinks this new algorithm is a major breakthrough in the field. He expressed optimism for the future, stating, “we continue to be optimistic that within five years, we will see real-world applications that are only possible with quantum computers.”
Addressing Complex Molecular Simulations
One of the most difficult endeavors for traditional supercomputers is modeling molecular interactions with precision. As the size of the molecule grows, so does the complexity. This, in turn, makes it difficult for classical systems to deliver trustworthy predictions. Google’s 54-qubit Sycamore quantum computer used for the calculation that took only 200 seconds. It predicted that Summit, the time world’s most powerful supercomputer at the time, would take 10,000 years to complete the same task.
Google’s 105-qubit Willow chip recently proved its mettle by finishing a benchmark computation in less than five minutes. By comparison, that same task was estimated to take Frontier, the fastest supercomputer at the time, an unimaginable 10 septillion years! This extreme contrast underscores the revolutionary power of quantum computing for scientific breakthrough.
Nobel-laureate Michel Devoret has explained, It’s not only the number of qubits that count. He explains how the promising low error rate—of around 0.1 percent on Willow—enabled the successful implementation of this algorithm. By accurately predicting the behavior of molecules through simulation, researchers are increasingly able to simulate complicated chemical problems.
The Path Forward and Implications for Future Applications
If quantum computing technology moves quickly enough, the potential implications for our molecular simulations are huge. Willow has now been able to successfully pair its new algorithm with molecule simulations through experiments of up to 15 qubits. This new development now enables researchers to build correct and accurate atomic-level models of molecular systems.
Even with these results, experts caution that what has been accomplished to date is still not yet “beyond classical.” We were inspired by this sharing from Thomas O’Brien, a staff research scientist at Google Quantum AI. He emphasizes the importance of data verification, noting, “If I can’t prove to you that the data is correct, how can I do anything with it?” This claim highlights the importance of strong validation practices as quantum technologies are used more widely in scientific discovery.
O’Brien highlighted that “the key aspect of verification is that it can lead to applications.” Researchers are still in the process of actively refining these algorithms and validating their outputs. Such a collaborative endeavor would open up exciting new doors for real-world applicability in areas such as drug discovery and materials science.