Beijing has reported early results from its first batch of non-consensus science innovative projects, a new funding mechanism to support research ideas that lack broad agreement among experts.

Launched in 2025 by the Beijing Natural Science Foundation, the initiative backs ideas that lack broad expert agreement but may carry transformative potential.

Seventy-two projects selected under the program have begun to show tangible outcomes, with several non-consensus ideas achieving breakthroughs, according to the Beijing Municipal Science and Technology Commission.

"Since the first batch of non-consensus projects was launched, overall progress has been good, with innovation vitality continuously unleashed. We have seen breakthroughs in areas such as solar-thermal hydrogen, new storage technologies, and brain-computer interfaces," said Liu Yang, deputy director of the general office of the Beijing Natural Science Foundation.

The non-consensus science innovation projects adopt a real-name expert-recommended review mechanism to support projects centered on major original and disruptive innovations.

Unlike traditional funding, the program allows a project to be approved if even a single expert gives recommendation, despite broader disagreement. Officials said the mechanism is intended to open a "green channel" for developing breakthrough innovation.

One of the early breakthroughs came from Tsinghua University, where researchers developed a light-and-heat co-catalysis pathway for hydrogen production.

The method improves yields under mild conditions and reduces carbon emissions by converting by-products into formaldehyde, offering a greener alternative to traditional high-temperature processes.

"Traditional hydrogen production requires high temperatures and generates large amounts of carbon dioxide. We wanted to change that, and we wanted to avoid such high temperatures. So, we first proposed a new concept, something that had no precedent before," said Tang Junwang, professor at the Department of Chemical Engineering of Tsinghua University.

The program adopts double-blind reviews, concealing the information of both reviewers and applicants to eliminate interference from nepotism. The real-name recommendation system requires experts to lend personal credibility to their endorsements, thus reinforcing accountability.

"The first thing I paid attention to is its disruptive technology, as the core value of non-consensus projects is to support disruptive innovation from zero to one. Of course, there may be risks, but non-consensus project initiative allows failure," said Tian Zhenyu, director of the Combustion Dynamics Research Center at the Chinese Academy of Sciences' Institute of Engineering Thermophysics.

The expert pool for the initiative includes more than 3,700 leading scholars, among them academicians and national-level talent, covering disciplines from mathematics and chemistry to medicine and engineering.