Invented by Movassagh; Ramis

The world is quickly moving toward quantum computing, but protecting sensitive information in this space is a real challenge. A new patent application proposes a clever way to keep quantum computations private—using something called “masked quantum circuits.” Let’s break down what this means, why it matters, and how this invention stands out in the industry.

Background and Market Context

Quantum computing is changing how we think about computers. Instead of using regular “bits” that are either 0 or 1, quantum computers use “qubits.” Qubits can be both 0 and 1 at the same time, thanks to something called superposition. This lets quantum computers solve some problems way faster than regular computers.

But here’s the thing: quantum computers are expensive and hard to build. Most people or companies who want to use a quantum computer have to send their problems to a quantum computing service provider—think of it like renting time on a supercomputer. If you’re a business or a researcher and you have a secret problem or algorithm, you might worry about sharing it with someone else’s machine. You don’t want your trade secrets or sensitive data to get out.

This need for privacy is growing. More companies want to take advantage of quantum computing, but they don’t want to risk giving away their secrets. This problem is not just about privacy for big businesses; it matters for anyone who has valuable data or ideas: scientists, medical researchers, banks, and even governments.

Today, if you want to keep your quantum computation private, your options are limited. You might try to simulate your quantum circuit on your own computer, but quantum simulations are slow and can become impossible as the problems get bigger. Some other solutions try to hide or scramble your data, but these are often too basic, too slow, or simply don’t scale well.

So, the market is hungry for a solution that lets you use powerful quantum computers without giving up your secrets. The patent application we’re discussing offers exactly that: a method to perform quantum computations on someone else’s quantum computer, but in a way that keeps your actual computation hidden. This is done through “masking” the quantum circuit, which means changing it just enough that the provider doesn’t know what you’re doing, but you can still get your answer.

If this approach works well and is easy to use, it could unlock new opportunities for quantum computing in finance, chemistry, machine learning, and many other fields. It could also encourage more companies and researchers to jump into quantum computing sooner, knowing that their data will remain private.

Scientific Rationale and Prior Art

To understand why this invention is important, let’s look at how quantum computers work and what has been tried before.

Quantum computers are built around the idea of quantum gates and qubits. You can think of quantum gates like instructions or actions that change the state of qubits. A quantum circuit is just a sequence of these gates. Each gate can be described by a mathematical object called a “unitary matrix.” When you apply a quantum circuit to qubits, you’re changing their state in a way that depends on the order and type of gates used. The final result is measured to get an answer to your computation.

If you want to use someone else’s quantum computer but don’t want them to know your circuit, you could try hiding the details. Traditional computer science has some methods for this, like homomorphic encryption. But these methods don’t work well for quantum circuits—they are too slow, or they don’t handle quantum data at all.

There are also some early attempts at “blind quantum computation.” These ideas let a user send a scrambled version of a quantum computation to a provider, who then computes the answer without knowing what the real computation was. But these early ideas often require a lot of extra resources or trust in the system. Some need the user to have special quantum hardware, or they only work for very simple circuits.

Another approach is to simulate the quantum circuit on a regular computer, keeping everything private. But this is only practical for very small circuits. Quantum computers are powerful exactly because they can handle much bigger problems—problems that would take a regular computer millions of years.

So, the challenge is clear: how can we let users send their quantum computations to a remote provider while keeping the real circuit secret? The answer needs to be fast, practical, and not require the user to own any quantum hardware. That’s where this patent application stands out.

The invention introduces a method for masking quantum circuits. The idea is to change some gates in the user’s quantum circuit in a controlled way, producing a new “masked” circuit. The quantum provider doesn’t know the original circuit but can still run the masked circuit. The user can then use the results to figure out the answer to their real problem.

Before this invention, there was no practical way to do this at scale. The approach in this patent cleverly uses mathematical tools like Cayley functions and rational functions to make the masking both efficient and hard to reverse. It also allows for checking the accuracy of the results, so the user can be sure the provider did the computation correctly.

This is more than just scrambling data; it’s about transforming the quantum circuit in a way that hides the original but still lets the user get the answer. The science behind this involves advanced math, but the result is a system that is both private and efficient.

Invention Description and Key Innovations

Now let’s dive into what this invention actually does and why it’s so clever.

The main idea is simple: take your real quantum circuit and create a new, “masked” version of it. The masked circuit is similar to the real one, but some of the quantum gates have been replaced or changed in a way that hides your real computation. You send this masked circuit to the quantum provider, who runs it and sends back the results. You then use your knowledge of the masking to translate the results back into the answer for your original circuit.

Here’s how it works, step by step:

First, you start with your original quantum circuit. This is just a list of quantum gates, each described by a unitary matrix. You pick some gates to mask—meaning you’ll change them so that the provider can’t tell what your real circuit is.

To mask a gate, you might use a random process to choose a new gate from all possible gates. This is called random sampling. Or, you might use a mathematical function—like a Cayley function—to create a smooth transformation from your real gate to a new, masked gate. The Cayley function is a special formula that lets you create a path between two gates. This path can be controlled by a real number, so you can create many different masked versions of your circuit.

Once you’ve created your masked circuit, you send it to the quantum provider. The provider runs the computation and sends back the results. These results are “masked,” meaning they don’t directly give away your real computation.

Here’s where the magic happens: you use the results from the masked circuit, along with your knowledge of the masking process, to reconstruct the answer to your real problem. This might involve using a rational function—a mathematical formula that describes how the outputs of the masked circuits relate to the output of your real circuit. If you need more precision, you can ask the provider to run the circuit more times, or to use different masked versions, until you’re satisfied with the accuracy.

This invention also lets you check if the provider did the computation right. You can simulate part of the masked circuit on your own computer and compare the results. If the provider is honest, the results should match up within a certain margin of error. If not, you know something’s wrong.

The key innovations in this invention are:

Masking Quantum Circuits in a Controlled Way: By choosing how and where to mask gates, the user can hide their real computation while still being able to get the answer. The masking can be done randomly or using mathematical functions that make it hard to reverse-engineer the original circuit.

Using Cayley and Rational Functions: These mathematical tools allow the user to create a wide range of masked circuits that are mathematically linked to the original. The Cayley function is especially good for smoothly transforming one gate into another, and rational functions are used to relate the results of the masked circuits back to the original answer.

Privacy-Preserving Delegation: The quantum provider never sees the real circuit, only the masked version. This means the user’s secrets are protected, even if the provider is untrusted or malicious.

Efficient Verification: The user can check the results from the provider by simulating part of the computation or by using known test cases. This builds trust in the system and makes it harder for the provider to cheat.

Scalable and Practical: The method is designed to be efficient, so it works even for large quantum circuits. Unlike previous ideas that only worked for small examples or required too much extra work, this approach is practical for real-world use.

In action, this system can be used by businesses, researchers, or anyone who needs to keep their quantum computations private. It could be built into quantum cloud services, letting users upload masked circuits and download results without ever revealing their secrets. This could open up quantum computing to more users and make it safer to use for sensitive tasks like finance, cryptography, or drug discovery.

The patent also covers different ways of implementing the system, including using special software, hardware, or even a mix of classical and quantum machines. The invention is flexible, designed to work with many types of quantum computers and circuits.

Conclusion

Quantum computing is a powerful new technology, but privacy concerns have held back its full potential. This patent application brings a fresh solution: a way to mask quantum circuits so users can safely delegate their computations to a quantum provider without revealing their secrets.

By combining clever mathematics with practical engineering, the invention allows users to create masked versions of their quantum circuits, send them to a remote quantum computer, and still recover the answer to their real problem. The provider never learns the details of the user’s computation. This is a big leap forward—making quantum computing safer, more private, and more useful for everyone.

If you’re interested in quantum computing but need to protect your sensitive data or trade secrets, this new approach could be a game changer. It’s a practical, scalable, and efficient way to bring the power of quantum computing to more users while keeping their secrets safe.

Click here https://ppubs.uspto.gov/pubwebapp/ and search 20250217688.