IBM and Google Are Also Making Moves in Quantum. Don’t Count Out the Bigger Players.

Researchers at International Business Machines ran a computation on IBM’s quantum hardware, then used hardware from Advanced Micro Devices to interpret the results. (COURTESY IBM)

Key Points

• IBM said it used AMD hardware to run a quantum error correction algorithm, advancing hybrid quantum systems.
• IBM’s stock increased by 7.5% following the announcement, marking its largest percentage gain since January.
• Google said its Willow quantum chip had been used to demonstrate the first algorithm to achieve verifiable quantum advantage, performing tasks beyond traditional computers.

Pure-play quantum-computing companies, the likes of Rigetti Computing, IonQ, and D-Wave Quantum, have basked in the spotlight this year. But larger, more established names in the tech sector are also making moves.

International Business Machines said Friday that it had used hardware from Advanced Micro Devices to run a quantum-error correction algorithm, representing a step toward building “practical and realistic” hybrid systems.

Shares of IBM climbed 7.5% following the announcement, putting the stock on pace for its largest percentage increase since January, according to Dow Jones Market Data. The stock has gained nearly 9% this week, its best weekly performance since the start of the year.

Rigetti, IonQ, and D-Wave gained 0.1%, 2.1%, and 5.8%, respectively.

Earlier this year, IBM unveiled a road map to building a fault-tolerant quantum computer, or a machine that can catch and correct errors as they occur. Part of this meant finding a way to decode superconducting qubits, the basic units of information in IBM’s machines, that could apply to increasingly large computers.

Decoding, in simple terms, is the process of interpreting data from numerous physical qubits to determine and correct errors. In the latest experiment, scientists ran a computation on IBM’s quantum hardware and used AMD FGPA, described as “a commonly available and relatively inexpensive classical compute engine,” to interpret the results.

The researchers said they were able to do so much faster than using costly graphics processing unit clusters, the technology typically used for training artificial-intelligence models and other high-performance computing tasks.

A paper containing these results was submitted to arXiv, a research-sharing platform from Cornell University, and is slated to publish Sunday evening. The paper isn’t peer reviewed.

The IBM announcement came just days after Alphabet Google said its Willow quantum chip had been used to demonstrate “the first-ever algorithm to achieve verifiable quantum advantage.” In simple terms, the company is saying its quantum system can perform tasks outside the reach of traditional computers.

On Monday, a peer-reviewed paper published in the journal Nature showed how a team of Google researchers repeatedly scrambled information before reversing the process. While it may not seem practical, the research helps scientists better understand the way information can be jumbled and disordered within a quantum system.

The authors of the paper were more restrained in their language, saying the results “indicate a viable path to practical quantum advantage.” And the experts who reviewed the work were split on whether the results demonstrated true quantum advantage.

Google is often credited with kicking off the quantum mania that is permeating the market. The launch of Willow at the end of 2024 was quickly followed by announcements from Microsoft and Amazon.com, who unveiled quantum chips of their own.

Write to Mackenzie Tatananni at mackenzie.tatananni@barrons.com