UNSW Sydney nano-tech startup Diraq has shown its quantum chips aren’t just lab-perfect prototypes—they also hold up in real-world production, maintaining the 99% accuracy needed to make quantum computers viable.

Diraq, a pioneer of silicon-based quantum computing, achieved this feat by teaming up with the European nanoelectronics institute Interuniversity Microelectronics Center (imec). Together they demonstrated the chips worked just as reliably coming off a semiconductor chip fabrication line as they do in the experimental conditions of a research lab at UNSW.

UNSW Engineering Professor Andrew Dzurak, who is the founder and CEO of Diraq, said up until now it hadn’t been proven that the processors’ lab-based fidelity—meaning accuracy in the quantum computing world—could be translated to a manufacturing setting.

“Now it’s clear that Diraq’s chips are fully compatible with manufacturing processes that have been around for decades.”

In a paper published in Nature, the teams report that Diraq-designed, imec-fabricated devices achieved over 99% fidelity in operations involving two quantum bits—or “qubits.”

The result is a crucial step toward Diraq’s quantum processors achieving utility scale, the point at which a quantum computer’s commercial value exceeds its operational cost. This is the key metric set out in the Quantum Benchmarking Initiative, a program run by the United States’ Defense Advanced Research Projects Agency (DARPA) to gauge whether Diraq and 17 other companies can reach this goal.

Utility-scale quantum computers are expected to be able to solve problems that are out of reach of the most advanced high-performance computers available today. But breaching the utility-scale threshold requires storing and manipulating quantum information in millions of qubits to overcome the errors associated with the fragile quantum state.

“Achieving utility scale in quantum computing hinges on finding a commercially viable way to produce high-fidelity quantum bits at scale,” said Prof. Dzurak.

“Diraq’s collaboration with imec makes it clear that silicon-based quantum computers can be built by leveraging the mature semiconductor industry, which opens a cost-effective pathway to chips containing millions of qubits while still maximizing fidelity.”

Silicon is emerging as the front-runner among materials being explored for quantum computers—it can pack millions of qubits onto a single chip and works seamlessly with today’s trillion-dollar microchip industry, making use of the methods that put billions of transistors onto modern computer chips.

Diraq has previously shown that qubits fabricated in an academic laboratory can achieve high fidelity when performing two-qubit logic gates, the basic building block of future quantum computers. However, it was unclear whether this fidelity could be reproduced in qubits manufactured in a semiconductor foundry environment.

“Our new findings demonstrate that Diraq’s silicon qubits can be fabricated using processes that are widely used in semiconductor foundries, meeting the threshold for fault tolerance in a way that is cost-effective and industry-compatible,” Prof. Dzurak said.

Diraq and imec previously showed that qubits manufactured using CMOS processes—the same technology used to build everyday computer chips—could perform single-qubit operations with 99.9% accuracy. But more complex operations using two qubits that are critical to achieving utility scale had not yet been demonstrated.

“This latest achievement clears the way for the development of a fully fault-tolerant, functional quantum computer that is more cost effective than any other qubit platform,” Prof. Dzurak said.

KAIST research team’s independently developed humanoid robot boasts world-class driving performance, reaching speeds of 12km/h, along with excellent stability, maintaining balance even with its eyes closed or on rough terrain. Furthermore, it can perform complex human-specific movements such as the duckwalk and moonwalk, drawing attention as a next-generation robot platform that can be utilized in actual industrial settings.

Professor Park Hae-won’s research team at the Humanoid Robot Research Center (HuboLab) of KAIST’s Department of Mechanical Engineering developed the lower body platform for a next-generation humanoid robot. The developed humanoid is characterized by its design tailored for human-centric environments, targeting a height (165cm) and weight (75kg) similar to that of a human.

The significance of the newly developed lower body platform is immense as the research team directly designed and manufactured all core components, including motors, reducers, and motor drivers. By securing key components that determine the performance of humanoid robots with their own technology, they have achieved technological independence in terms of hardware.

In addition, the research team trained an AI controller through a self-developed reinforcement learning algorithm in a virtual environment, successfully applied it to real-world environments by overcoming the Sim-to-Real Gap, thereby securing technological independence in terms of algorithms as well.

Currently, the developed humanoid can run at a maximum speed of 3.25m/s (approximately 12km/h) on flat ground and has a step-climbing capability of over 30cm (a performance indicator showing how high a curb, stairs, or obstacle can be overcome). The team plans to further enhance its performance, aiming for a driving speed of 4.0m/s (approximately 14km/h), ladder climbing, and over 40cm step-climbing capability.

Professor Hae-Won Park’s team is collaborating with Professor Jae-min Hwangbo’s team (arms) from KAIST’s Department of Mechanical Engineering, Professor Sangbae Kim’s team (hands) from MIT, Professor Hyun Myung’s team (localization and navigation) from KAIST’s Department of Electrical Engineering, and Professor Jae-hwan Lim’s team (vision-based manipulation intelligence) from KAIST’s Kim Jaechul AI Graduate School to implement a complete humanoid hardware with an upper body and AI.

Through this, they are developing technology to enable the robot to perform complex tasks such as carrying heavy objects, operating valves, cranks, and door handles, and simultaneously walking and manipulating when pushing carts or climbing ladders. The ultimate goal is to secure versatile physical abilities to respond to the complex demands of actual industrial sites.

During this process, the research team also developed a single-leg “hopping” robot. This robot demonstrated high-level movements, maintaining balance on one leg and repeatedly hopping, and even exhibited extreme athletic abilities such as a 360-degree somersault.

Especially in a situation where imitation learning was impossible due to the absence of a biological reference model, the research team achieved significant results by implementing an AI controller through reinforcement learning that optimizes the center of mass velocity while reducing landing impact.

Professor Park Hae-won stated, “This achievement is an important milestone that has achieved independence in both hardware and software aspects of humanoid research by securing core components and AI controllers with our own technology.

“We will further develop it into a complete humanoid, including an upper body to solve the complex demands of actual industrial sites and furthermore, foster it as a next-generation robot that can work alongside humans.”

The results of this research will be presented by JongHun Choe, a Ph.D. candidate in Mechanical Engineering, as the first author, on hardware development at Humanoids 2025, an international humanoid robot specialized conference held on October 1st.

Additionally, Ph.D. candidates Dongyun Kang, Gijeong Kim, and JongHun Choe from Mechanical Engineering will present the AI algorithm achievements as co-first authors at CoRL 2025, the top conference in robot intelligence, held on September 29th.

The presentation papers are available on the arXiv preprint server.

If the UK’s digital economy is to enjoy a prosperous future, the nation’s electricity grids need to be urgently upgraded to ensure the datacentres underpinning it all have a consistent and reliable supply of high-capacity power.

That’s according to a poll of 101 FTSE 250 executives, commissioned by the Energy Networks Association (ENA), where 90% of participants said they believe that upgrading the grid is “essential to unlocking high-growth industries”, while more than 80% said the UK cannot compete globally without these upgrades.

“This requirement is particularly acute for datacentres, which underpin AI [artificial intelligence] model training, cloud computing and the broader tech ecosystem,” said the ENA, in a statement.

Of those polled, 55% said they are confident the UK can become a global leader in AI, for example, but realising this ambition hinges on the availability of reliable, high-capacity power.

Furthermore, 75% of those questioned said planned grid improvements would enable their business to invest and grow in the UK, with just 19% stating that they think the current grid is sufficient to meet the future energy demands of high-growth industries.

“This polling shows a high level of confidence among UK business leaders in our ability to lead the world in clean energy, advanced manufacturing, life sciences and digital technology,” said Energy Networks Association CEO Lawrence Slade. “They see the UK’s potential to dominate the industries of tomorrow, and the benefits this would bring for jobs, investment and growth,” he said. “But they are equally clear that this opportunity is not guaranteed. It hinges entirely on a future-ready electricity grid.

“Without it, we risk losing our competitive edge and missing out on the economic, environmental and social gains that are within our grasp. Decisions made this year will be pivotal for securing the private investment network operators need to make this future-ready grid a reality.”

Concerns about the UK grid’s ability to withstand the number of datacentres being planned and built have been circulating for years, but have ramped up more recently in the wake of the government’s rhetoric about wanting to become an AI superpower.

In line with this, there has been a flurry of announcements from datacentre developers about their plans to build hyperscale, high-performance compute and AI server farms across the country, which typically consume more energy than a traditional datacentre.

According to the UK government’s own figures, domestic datacentre capacity could rise to between 3.3GW and 6.3GW by 2030, mainly due to the growing demand for AI workloads, which will place further pressure on the grid.

Teodora Kaneva, head of smart infrastructure and systems at TechUK, said the government has acknowledged the pressure these datacentre developments will put on the grid, and allocated “significant investment” in the nation’s energy infrastructure as a result.

“Now we must focus on how we can deliver at pace to deliver on national ambition, [because] delays in power provision prevent the construction of new digital infrastructure and science parks, driving investment elsewhere,” she said.

“This is not only an issue for the UK tech sector itself, but it also affects the chancellor’s AI and life sciences ambitions. Reliable, affordable and flexible power is essential for the government’s growth mission.

“With an increasingly polarised debate on infrastructure delivery, politicians from all parties must come together to champion this vital national infrastructure,” said Kaneva.