AI-rewritten: This is a summary of an article from New Scientist, rewritten by AI (Qwen, running locally) to make it easier to read. The facts come from the original article – read it for the full story.

---

Google Quantum AI researchers have developed a method to fix errors in quantum computers more efficiently, potentially making future machines more useful than expected. While quantum computers can perform complex mathematical tasks that standard devices cannot, they are currently very error-prone. To work reliably, their components must be grouped into "logical qubits," but running specific operations called non-Clifford gates traditionally requires huge numbers of physical qubits. This creates a major engineering hurdle for building practical machines.

Emma Rosenfeld and her team used Google’s Willow quantum computer to demonstrate a new technique that reduces errors by 40 times. The core of this advance is "magic state cultivation," which allows the computer to generate special quantum states needed for error correction within smaller groups of qubits rather than combining many imperfect ones. In theory, this approach could allow quantum computers to run important programs with just one-tenth the number of qubits previously thought necessary.

Experts note that while this method works similarly to experiments using ion-based qubits, Willow’s superconducting circuits may reduce overall computation time. However, challenges remain; in the test, only 8% of runs produced high-quality magic states. Researchers like Isaac Kim from UC Davis suggest the technique could make fault-tolerant computing cheaper, but Zhenyu Cai from Imperial College London warns that error rates must drop further before tackling truly useful problems. Rosenfeld added that success rates should improve as hardware refines and the benchmarking method evolves.