From October 20-22 by the Huangpu River in Shanghai, the 2026 Sustainable Global Leaders Conference returns once again. Nobel and Turing Award winners will share the future and trends of technology on site, along with renowned professors who "rescue" people from anxiety and the era of involution—McFarlane and Xiang Biao in a live dialogue!
Limited spots available, first come first served. Let us journey together toward sustainability! [Click to Register]
Alphabet announced that Alphabet's Quantum AI team has successfully broken through to resolve a key bottleneck in quantum computing: it has empirically demonstrated the feasibility of a "magic state cultivation" scheme on the Willow processor, achieving approximately a 40-fold reduction in errors compared to not using this technology. This breakthrough achievement was published today in the journal Science, marking a critical step toward scaling quantum computing for practical use.
The Magic State Bottleneck
As early as 2004, long before quantum computing became an engineering practice direction, researchers foresaw a thorny bottleneck. To achieve universal quantum computing and run programs with practical value, it is necessary to be able to execute a specific set of quantum operations. The way to achieve this is to first prepare so-called "magic states," named because such quantum states can generate the quantum operations needed to surpass classical computation. But here lies a difficulty: preparing high-quality, error-corrected magic states is extremely resource-intensive. Some estimates show that a quantum computer might need to use up to 95% of its total computing power to generate magic states, leaving very few resources for actual computation. There is another complication: once a magic state is used, it cannot be reused—it is a scarce resource. Therefore, for two decades, researchers have been continuously working to reduce the resource overhead of magic state preparation.
How This Breakthrough Was Achieved
In 2024, Alphabet Quantum AI researchers Craig Gidney, Noah Shutty, and Cody Jones drew on ideas from the literature and optimized them, developing what is now called magic state cultivation technology. This scheme is expected to significantly reduce resource consumption. The core idea is: the cultivation process acts like a filter. Noisy quantum states are injected into a small region of the quantum chip, followed by a series of highly coordinated measurements to detect errors and eliminate unqualified samples. Craig's team proved that with this method, quantum states of extremely high purity and low error can be "cultivated," and can be simultaneously converted into a larger-scale, practically valuable encoded form. This scheme performs excellently in theory. If it can be implemented on quantum hardware, it could greatly reduce the scale of quantum computers needed for practical tasks.
Like many new technology developments, engineering implementation requires a series of technological innovations and engineering challenges to be overcome. For example: during the first test of the cultivation protocol, Alphabet Quantum AI researcher Emma Rosenfeld observed data that appeared to violate physical laws. After months of investigation, simulation, peer discussions, and consistency verification, she discovered that the problem stemmed from erroneous assumptions in the team's commonly used measurement methods. Specifically, a type of noise caused by drift in laboratory electrical equipment and temperature fluctuations interfered with the validity of benchmark test results. The team needed to simultaneously adjust the theoretical framework and engineering solutions. The research team added extra control measures to the system to weaken the impact of this noise; they also innovated a benchmarking method that uses the cultivation technology itself as a measurement tool, solving the problem at its root.
Building on this achievement, along with more algorithmic innovations and engineering solutions, Alphabet has pushed quantum metrology technology to a new pinnacle. On Alphabet Quantum AI's Willow processor, this protocol achieved up to a 40-fold error compression compared to quantum states without cultivation processing; the resulting T-state logical fidelity reached a record-breaking 99.99%. This achievement has now been published in Science, empirically demonstrating for the first time a feasible solution to one of quantum computing's most severe resource bottlenecks, proving on real hardware that Alphabet can efficiently prepare this critical magic state.
Research Findings and Future Outlook
Alphabet also preliminarily verified that the theory proposed by Craig and others can be adapted to its quantum computing architecture. The team achieved switching between "color codes" (a new type of quantum error correction code, highly suitable for magic state cultivation) and "surface codes" (a more mature error correction scheme), grafting a distance-3 color code quantum state onto a distance-5 surface code. This grafting protocol first originated from a 2024 theoretical paper. This experiment confirmed that the purified quantum states can stably persist and be smoothly transferred within the hardware.
Comments