Researchers unveil self-testing quantum random chip

Researchers from Squareroot8 Technologies and partner institutions have published what they describe as the first fully integrated self-testing quantum random number generator on a silicon photonic chip. The work appears in PRX Quantum.

The chip, called QDice, is designed to generate random numbers while checking its own integrity during operation. It uses a measurement-device-independent protocol that treats measurement hardware as untrusted and continuously tests it, producing certified random bits only when the device meets a required threshold.

Random number generation underpins digital security systems used in banking, cloud computing, communications networks and other connected infrastructure. A longstanding challenge has been proving that output remains secure over time as hardware ages, drifts or is tampered with.

Most random number generators in use today rely on the assumption that hardware was correctly characterised at manufacture and continues to behave as expected throughout its working life. The researchers argue that this creates a security gap because predictable output may not be detected by existing verification methods.

The chip design

QDice is built on a single silicon photonic chip fabricated at Singapore's Advanced Micro Foundry. It operates at room temperature and does not require cryogenic cooling, unlike some other quantum systems that depend on more complex operating environments.

The chip was made using a standard eight-inch silicon-on-insulator wafer process, a manufacturing platform already used in data centre and telecommunications hardware. This gives the work relevance beyond the laboratory by pointing to a route through established semiconductor production lines.

According to the researchers, the system reaches a higher security level than conventional fully trusted quantum random number generators because its security analysis assumes that a threat actor could hold quantum correlations with the chip and that everything except the quantum light source may be untrusted. The team says this framework has no direct classical equivalent.

The trade-off is speed. A typical quantum random number generator may run at more than 100 gigabits per second, while QDice currently produces random numbers at 64 bits per second.

Commercial question

That gap matters if the technology is to move into mainstream security products. The team says it has fabricated photodiodes that would lift detector efficiency from 69 per cent to 92.4 per cent, increasing output to 68 megabits per second.

If achieved in practice, that would move the device closer to rates that could support commercial deployment in applications where continuous verification of randomness matters more than raw speed. Potential uses mentioned by the researchers include financial systems, government networks, cloud infrastructure, medical devices and connected hardware.

The project brought together researchers from quantum technology, banking and financial services, commercial technology and academic institutions. That mix reflects broad interest in cryptographic systems that can provide stronger assurances about the quality of the randomness they produce.

For organisations in heavily regulated sectors, the issue is not simply whether random numbers can be generated, but whether they can be shown to remain private and unpredictable as components degrade or operating conditions change. In that sense, the work addresses a practical cyber security question, not only a theoretical one in quantum science.

Dr Ng Hong Jie described where the device fits in the market.

"QDice sits in a very sweet spot between the security versus practicality trade-off, making it one of the most secure QRNGs that are commercially available in the market," said Dr Ng Hong Jie, Senior Software Engineer, Squareroot8 Technologies.

Dr Goh Koon Tong said the work could help bring quantum security tools into wider use.

"Random numbers play a central role in modern security and digital trust. Quantum theory allows us to check for signatures in these numbers to verify their secrecy. What we have done here is turn quantum light into verifiably secure random numbers with the entirety of the process within a single chip smaller than the size of your fingernail. This will undoubtedly be a major step towards mass adoption of quantum technologies," said Dr Goh Koon Tong, Co-Founder and Chief Technologist, Squareroot8 Technologies.