RIKEN proposes fewer steps to prepare protected quantum bits
A newly published simulation study tackles the extra work needed before error-corrected quantum computing can begin.
Reading tools for text size, copying, printing and your reading list

RIKEN researchers have designed a less resource-intensive way to prepare protected quantum information, reducing the initial preparation circuit from 108 two-qubit operations to 67, before additional error-checking steps. RIKEN announced the work on November 14, following publication in Science Advances on November 13. The result is a circuit design tested in simulations, not a newly operating quantum computer.[1]
The study addresses a practical catch in . Spreading information across several can protect it against errors, creating what researchers call . But preparing those protected states also takes operations that can introduce errors.[1][2]
Naoyuki Kanomata and Hayato Goto studied a code that places six logical qubits in a block of 30 physical qubits. The shortened circuit prepares a specific starting state, called the all-plus state. Separate verification checks then catch dangerous errors. Those checks require additional qubits: the 30-qubit block is not the total size of the preparation system.[2]
RIKEN reports that the proposed approach reduces the average total number of physical qubits needed for state preparation to half or less of a comparison method using six copies of the encoded state, while maintaining similar simulated error rates for logical two-qubit operations. That comparison concerns state preparation, not the total resources of an entire useful quantum computation.[1]
The core idea was already public in a preprint submitted on September 18, 2025. This week's news is the journal publication and accompanying RIKEN announcement, rather than the first disclosure of the approach.[1][3]
The earlier manuscript's simulations omit memory errors and single-qubit-gate errors. A hardware experiment would therefore need to establish how well the savings survive a device's actual noise and operating constraints.[2]
Keep your reading trail
Keep up with these topics
Explore the subjects covered in this story.