Quantum Wars: Google, Microsoft, and Amazon’s Competing Paths to Fault-Tolerant Qubits

Following Google’s Willow in December 2024, we now have two additional new quantum chips announced in the final week of February 2025: Microsoft’s Majorana 1, and now Amazon’s Ocelot.

“AWS researchers,” claims Amazon, “have, for the first time, combined cat qubit technology and additional quantum error correction components onto a microchip that can be manufactured in a scalable fashion using processes borrowed from the microelectronics industry.”

Three new quantum chips from three separate quantum manufacturers in the space of three short months is a dramatic development. They all seek to solve the fundamental problem in quantum computing: the fragility of the quantum state is easily disrupted by the slightest environmental disturbance. The result is an intolerable level of processing errors.

Any solution requires complex quantum error correction, which is currently considered the single biggest roadblock to developing a usable quantum computer – the traditional approach (if anything about quantum can be called traditional) is to increase the number of qubits.

The problem here is the sheer number of error-correcting qubits (which we can call physical qubits) that are needed to provide one working qubit (which we can call the logical qubit). In essence, this requires sharing information across many – very many – physical qubits to realize one logical qubit. It is the huge number of physical qubits necessary that hinders the development of scalable quantum computers. Error correction is a major part of the solution by greatly reducing the number of necessary physical qubits.

“Quantum computing isn’t a one-size-fits-all technology,” comments Rebecca Krauthamer, founder and CEO at QuSecure. “There are multiple ways to build a quantum computer, each with different trade-offs.” Each of these new chips has taken a different approach to reducing quantum error; so, before looking at how Ocelot attempts to solve the quantum error problem, we should summarize the Willow and Majorana 1 approaches and compare them.

Willow

Using superconducting qubits, comments Troy Nelson, CTO at Lastwall, “Willow has been able to reduce the error rate of both bit-flip and phase-flip errors by improving the way physical qubit measurements are taken.”

Marc Manzano, general manager of cybersecurity at SandboxAQ, adds, “With its focus on improved qubit coherence and reduced error rates, Google’s Willow aims to reduce errors as the number of qubits scales.”

Advertisement. Scroll to continue reading.

Google confirmed this in its December announcement: “Using our latest advances in quantum error correction, we were able to cut the error rate in half. In other words, we achieved an exponential reduction in the error rate.”

It claims to have achieved ‘below threshold’ for its error corrections, where ‘threshold’ is a theoretical level for error rates. Being below threshold means it is possible to build arbitrarily large quantum computers.

For more on the Willow chip, see Google’s Willow Chip Signals the Urgency of Post-Quantum Cryptography Migration.

Majorana 1

Microsoft’s Majorana 1 does not use superconducting qubits. “It pursues a fundamentally different path to building qubits with topological qubits,” explains Krauthammer, “aiming for a more stable and fault-tolerant system by using exotic quantum states called Majorana zero modes. This was only theorized until Microsoft revealed the Majorana 1.”

More stable qubits imply fewer physical qubits per logical qubit will be required. Microsoft claimed that “Today, the company has placed eight topological qubits on a chip designed to scale to one million.” But this scalability is not yet proven in practice.

Scott Best, senior director of silicon security products at Rambus, adds, “The Majorana 1 chip is a notable breakthrough in quantum computing, but only time will tell if they’ve discovered a more feasible path to achieve the highly sought after ‘qubit scalability’.”

For more on the Majorana 1 and topological qubits, see What Microsoft’s Majorana 1 Chip Means for Quantum Decryption.

Enter the Ocelot. Architecturally, it is closer to Willow than it is to Majorana 1. “Both Ocelot and Willow rely on superconducting circuits and microwave control to manipulate qubits, and both face the same fundamental issue: superconducting qubits are fast and one of the more widespread architectures – but can be error-prone,” explains Krauthammer.

But Ocelot’s qubits come with a twist. Ocelot uses what are known as ‘cat qubits’, so named after Schrodinger’s thought experiment. Cat qubits are significant for bosonic error correction which encodes quantum information within the infinite-dimensional Hilbert space of a quantum harmonic oscillator.

This is further explained by Fernando Brandao and Oskar Painter in an Amazon blog: “One type of bosonic quantum error correction uses cat qubits, named after the dead/alive Schrodinger cat of Erwin Schrodinger’s famous thought experiment. Cat qubits use the quantum superposition of classical-like states of well-defined amplitude and phase to encode a qubit’s worth of information…

“A major advantage of cat qubits is their inherent protection against bit-flip errors. Increasing the number of photons in the oscillator can make the rate of the bit-flip errors exponentially small. This means that instead of increasing qubit count, we can simply increase the energy of an oscillator, making error correction far more efficient.”

“Willow has been able to reduce the error rate of both bit-flip and phase-flip errors by improving the way physical qubit measurements are taken,” says Nelson. “Ocelot, on the other hand, has designed its qubits to intrinsically suppress bit-flip errors. Known as ‘cat qubits’, this design has demonstrated up to 90% increase in quantum error correction efficiency.”

In this latter sense, Ocelot is closer to Majorana 1 – these chips attempt to reduce errors by qubit design rather than simply correct them.

“While Willow and Ocelot are refining existing technology, Majorana 1 is still in an earlier experimental stage, needing exotic materials and precise conditions that have yet to be proven scalable,” says Krauthammer. “The question is whether improving superconducting architectures will get us to fault-tolerant quantum computing faster, or if a fundamentally different qubit design is necessary to truly break through… Either way, all these approaches are pushing the field in a truly exciting way, toward the same goal: a practical, scalable quantum computer.”

Manzano adds, “These three chips represent distinct approaches to tackling the challenges of quantum computing. It’s still early days, and the ‘winning’ technology is yet to be determined. However, each advancement pushes the field forward and brings us closer to realizing the transformative potential of quantum computing.”