Google’s Quantumania ⚛
Alphabet’s wager on the future of computing.
Introduction
Last month, whilst I was writing up the downside of Google’s DOJ trial and the company’s potential break up, Alphabet announced their quantum processing chip, Willow. The announcement caused Google’s GOOG 0.00%↑ share price to jump by 5%.
You can watch the announcement video here: -
Alphabet has reported that the Willow chip only takes 5 minutes to solve a problem that would take the world's fastest super computers ten septillion – or 10,000,000,000,000,000,000,000,000 years – to complete.
Willow operates with 105 qubits and addresses the critical challenge of error reduction in quantum computing. By introducing more qubits and correcting errors in real time, Google has achieved what’s known as “below threshold” performance - a breakthrough in error correction first theorized in 1995.
What is Quantum Computing?
Before we get into the weeds (or willows🤦) it’s important to understand how classical computing and quantum computing differ.
Classical computing is like walking through a maze step by step to find the exit, while quantum computing is closer to exploring all paths in the maze at the same time.
Classical computing processes information sequentially using binary states (0s and 1s), while quantum computing uses quantum bits (qubits) that exist in multiple states at the simultaneously, enabling parallel processing.
Eight units of bits are referred to as one byte. Classical computers have less compute power and typically only return one result - true or false, yes or no - because bits of 1s and 0s are binary. Classical computers use serial processing, which is successive in nature, meaning one operation must be complete before another one follows.
With the explosion of AI, numerous computing chips use parallel processing, an expansion of classical processing, which can perform simultaneous computing tasks. Nvidia’s GPUs (Graphics Processing Unit) are the flagship examples of parallel processing. However, these examples are still closer to classical than quantum.
Quantum computers work in a fundamentally different way by harnessing quantum mechanics to solve problems faster. A qubit, unlike a classical bit, can exist in multiple states at once. This is due to a quantum phenomenon known as superposition. Consider Schrodinger's famous cat, which is both alive and dead at the same time, until observed. This allows quantum computers to process a vast number of possibilities all at once.
Quantum computers typically operate under highly regulated physical conditions. In order to maintain the quantum nature of the system, the processors are kept at temperatures close to absolute zero and isolated from external electrical, magnetic and thermal noise.
It's anticipated that quantum computers will eventually be able to vastly speed up complex processes that classical computers are incapable of.
Qubit vs Bit
A qubit, or quantum bit, is the basic unit of information in quantum computing. It's analogous to a bit in a classical computer, but qubits can store considerably more information than a standard bit leading to vastly superior computing power.
A qubit uses the quantum mechanical phenomena of superposition, achieving multiple possible states. A classical binary bit can only return one of two outcomes, such as 0 or 1, meaning that it can only be in one of two possible states.
A qubit cab represent 0, 1, or any proportion between 0 and 1 in superposition of both states, with a probability of being a 0 and a probability of being a 1.
Superposition allows quantum algorithms to process information in a fraction of the time it would take even the fastest classical systems so solve complex problems.
The amount of information a qubit system can represent grows exponentially, 500 qubits can achieve computations which would not be possible event with 2^500 classical bits.
It would take a classical computer millions of years to find the prime factors of a 2,048-bit number, but qubits can perform the calculation in just minutes.
There are numerous physical implementations of qubits. Where classical computers use familiar silicon-based chips, qubits can be made from trapped ions, photons, artificial or real atoms or quasiparticles.
Superposition: A qubit can represent a 0, a 1, or any proportion of both, with a certain probability of being each. This allows quantum computers to process information faster than classical computers.
Entanglement: Qubits can become entangled with other qubits. When they become entangled, the state of one qubit becomes linked to the state of another, no matter the distance between them. It's like having two magic dice that always land on the same number, even if they're rolled on opposite sides of the universe. This allows quantum computers to process information in a highly interconnected way.
Incompatible measurements: Qubits can be subjected to incompatible measurements. In quantum mechanics, incompatible measurements cannot be performed on a single copy of a quantum system at the same time. This is because measuring a system is an invasive process that disturbs the state.
Decoherence: When qubits lose their quantum state due to environmental interference this is know as Decoherence. This occurs when a quantum system interacts with its environment, such as through electromagnetic fields, temperature fluctuations, or measurement.
What is Willow?
Willow is Google’s latest quantum processor. The chip features 105 qubits - about double the number on Google's previous Sycamore chip. Willow was manufactured in Google's purpose-built manufacturing plant in California.
Google reported that Willow performed a standard benchmark calculation in under five minutes, that would take one of today’s fastest supercomputers 10 septillion (that is, 1025) years, a number that vastly exceeds the age of the Universe.
The test Google uses is specifically designed to test quantum systems. It's called the random circuit sampling (RCS) benchmark, which Google researchers devised several years ago and has since become standard in the field.
Errors are one of the greatest challenges in quantum computing, as qubits rapidly interact with the external environment. This makes it difficult to protect the information needed to complete a computation.
Typically the more qubits that are used, the more errors will occur, resulting in the loss of quantum nature. Google uses devices called cryostats to keep quantum chips temperature near absolute zero to limit information exchange with the environment.
Willow's design could exponentially reduce these errors as it scales up qubits. This feature is key as no previous system has come close to achieving. The more qubits that are used in Willow, the more errors are reduced, and the more quantum the system becomes. This occurs as the larger a surface of the code lattice, the more errors it can tolerate.
Even when a quantum computer is properly isolated and running, the qubit output is a probabilistic measurement of a classical bit, meaning it can still be wrong. In general, quantum computers produce more errors the more qubits they have, but Google has reportedly overcome that challenge with Willow, which has 105 qubits.
The expectation is that as the lattice gets bigger, the qubit is more and more protected and the performance improves. This has allowed Google so solve the key challenge of error correction which has plagued the quantum field for almost 30 years.
Google tested arrays of 3x3 qubits, scaling up to 5x5 and 7x7—each time, the error rate dropped (see below).
Willow addressed decoherence through advanced error correction techniques. By engineering qubits with longer coherence times and optimising how they interact, Willow minimises the effects of environmental disturbances.
Willow’s scalable error correction system allows for continuous detection and correction of decoherence-induced errors, ensuring computations remain accurate even as the number of qubits grows. This approach stabilises the system and enables it to achieve unprecedented levels of performance, making quantum computing a more viable solution for complex for real-world applications.
By mitigating decoherence, Willow enables more stable and reliable quantum computations for longer periods. This breakthrough allows Willow to perform more complex calculations with greater accuracy, marking significant progress in the quantum computing field.
Willow will provide computation power that can solve real-world problems traditional computers cannot. Potential applications include improving AI training data, accelerating drug discovery, designing more efficient batteries for electric vehicles and accelerating progress in new energy alternatives.
Hartmut Neven, who leads Google's Quantum AI lab that created the chip, told the BBC that Willow would be used in some practical applications. However, Neven advised that a chip able to perform commercial applications would not appear before the end of the decade.
The Quantum Computing Landscape
Outside of Google’s Willow, the quantum computing landscape is rapidly evolving with well known companies like Microsoft, Intel and IBM and start-ups including IonQ.
IBM: IBM’s QLDPC code leverages a unique connectivity strategy to minimize qubit overhead.
Microsoft: Through the Azure Quantum platform, Microsoft is developing a topological quantum computer with qubits that are inherently resistant to error. In 2024, Microsoft applied a qubit virtualisation system to Quantinuum's trapped ion quantum computer to create 12 logical qubits, demonstrating significant advancements in error correction and scalability.
Quantinuum: Formed from the merger of Honeywell Quantum Solutions and Cambridge Quantum Computing, Quantinuum has achieved milestones such as reaching a quantum volume of 1,048,576 with its H-Series systems. The company also holds the record for two-qubit gate fidelity, becoming the first to reach 99.9%.
Zuchongzhi 3.0 (China): Developed by a team led by Pan Jianwei at the University of Science and Technology of China. Zuchongzhi 3.0 is a quantum computer that has demonstrated performance comparable to Google's Willow. The system shows high precision in qubit operations and stability, with plans to incorporate advanced error correction techniques in the near future.
D-Wave Quantum: Specialising in quantum annealing technology, D-Wave focuses on optimization problems and has been a significant player in the quantum computing industry. The company has seen a substantial increase in stock value following recent advancements in the field.
IonQ: Utilizing trapped-ion technology, IonQ develops quantum computers aimed at solving complex problems in machine learning, chemistry and optimisation. The company is actively contributing to the advancement of quantum algorithms and applications.
Conclusion
Willow is, for now, an experimental device, meaning real world quantum computers remain years and billions of dollars away.
Quantum computers are still in the early stages, facing major technical challenges including maintaining quantum coherence, as qubits quickly lose their quantum state.
Willow is not just a technological breakthrough but a moonshot aimed at the future of computing. According to Google, Willow is much faster than previous designs, but the real advancement is its accuracy.
For nearly 30 years, the primary challenge in quantum computing has been reliability. As previously highlighted, as quantum systems scale with more qubits, they become increasingly error-prone.
Willow's design overcomes this by exponentially reducing errors as qubits scale - a breakthrough no previous system has achieved. Rather than simply adding more qubits, Willow enhances their reliability.
Willow has potentially made quantum chips infinitely more efficient. If the technology can be scaled, the application will support areas that require considerable computational power. Google has claimed a multitude of uses for Willow, from AI to drug discovery, climate modelling and problems that cannot be solved using classical computing.
Google’s announcement is unlikely to be a coincidence, published less than a month after the DOJ proposed that Alphabet should sell of Chrome and Android. Google is in desperate need of some positive kudos.
The US will want to lead in quantum technology and gain significant advantages in critical sectors like military and intelligence. The Willow chip could solidify the US's position as the global technology leader in this field and encourage the incoming President to take a lenient view of the DOJs ruling.
References and further reading
https://blog.google/technology/research/google-willow-quantum-chip/
https://www.techtarget.com/searchdatacenter/tip/Classical-vs-quantum-computing-What-are-the-differences
https://azure.microsoft.com/en-us/resources/cloud-computing-dictionary/what-is-a-qubit
https://www.forbes.com/sites/timothypapandreou/2024/12/16/googles-code-cracking-quantum-leap-heres-what-it-mean-for-business/
https://www.bbc.co.uk/news/articles/c791ng0zvl3o
https://www.wsj.com/articles/the-age-of-quantum-software-has-already-started-854
https://www.extremetech.com/computing/google-reveals-willow-quantum-computing-chip
https://www.marketwatch.com/story/ibms-stock-has-missed-the-quantum-rally-but-could-still-be-your-ticket-to-ride-that-hot-trend
https://www.barrons.com/articles/quantum-stocks-willow-google-
https://www.wsj.com/articles/the-age-of-quantum-software-has-already-started
https://thequantuminsider.com/2024/12/28/googles-quantum-error-correction-has-some-competition/
This is an impressive leap forward for quantum computing. Willow’s ability to tackle error correction and scale up qubits is a game-changer. It’s crazy to think that we’re talking about something that could solve problems in minutes that would take current supercomputers millions of years! Still, we’re a few years away from real-world applications, but it’s exciting to see how fast the tech is advancing. Definitely something to keep an eye on in the coming years.