AIThis post was created with the assistance of artificial intelligence (AI).

TL;DR

Google Quantum AI researchers and collaborators report that the Willow processor beat the best possible classical winning rate in two cooperative games where players could not communicate. The team also ran a 105-qubit mathematical test, but the classical comparison at that scale depends on an unproven conjecture.

Researchers from Google Quantum AI and collaborators report that the company’s Willow quantum processor beat classical limits in two cooperative games whose players were not allowed to communicate. The results, published in Science, offer a test of quantum contextuality and a way to probe how quantum effects hold up as more qubits are used; a separate 105-qubit experiment showed a valid result more than half the time, but does not establish a confirmed classical advantage at that scale.

Before you orderOffer from Amazon

Get the latest gadgets delivered free with Prime

  • Fast, free delivery on millions of items
  • Prime Video, Amazon Music and more included
  • Member-only deals all year
Start your free Prime trial Free trial for eligible customers · Cancel anytime
As an affiliate, we earn on qualifying purchases.

In the first test, two players received separate assignments based on a three-by-three grid: one was asked to fill a row and the other a column. Their answers had to meet different parity rules and match at the shared square. The players could agree on a strategy in advance but could not communicate after receiving their assignments. The rules cap the best classical strategy at a 88.9% win rate. Willow won 98.3% of the time across the test, according to the study.

The researchers ran all nine row-and-column combinations 100,000 times each. They also measured contextuality using a related test: the processor scored 5.59, against a classical ceiling of 4 and an ideal quantum score of 6. The team attributed the gap between the measured and ideal results to experimental errors.

A second game tested scaling across more qubits. The team reported winning above the classical limit with up to 71 qubits, each representing a player in the game. Separately, it verified genuine group entanglement among as many as 45 qubits. These are different measures: the 71-qubit result describes the scale at which the game still exceeded its classical winning rate, while the 45-qubit figure refers to verified shared entanglement.

At a glance
reportWhen: Study reported November 19, 2026; publi…
The developmentA study published in Science reports that Google’s Willow quantum processor exceeded precise classical limits in two no-communication games and tested a mathematical problem at up to 105 qubits.

How the Games Test Quantum Advantage

The two games provide a relatively direct way to compare quantum and classical strategies because their rules specify a precise classical ceiling. Beating that ceiling shows that the measured device performed beyond what the best permitted classical strategy can achieve under those rules. It does not by itself show that Willow is faster or more useful than classical computers for ordinary tasks.

The results also matter as a potential tool for hardware evaluation. The researchers say contextuality-based tests may help engineers identify errors that other checks could miss. As quantum processors grow, distinguishing useful quantum behavior from noise and implementation faults remains a central engineering challenge. The reported gap between Willow’s score and the ideal quantum result underscores that the hardware is not error-free.

The 105-qubit test is a more tentative step toward a computational advantage. Willow returned a valid answer to a two-dimensional hidden linear function problem more than half the time. However, the estimate of the classical computing effort needed for that problem relies on a conjecture. The study therefore does not establish a measured, definitive advantage over classical computing for that experiment.

Products Worth Considering

Logitech MX Creative Console with 9 Customizable Keys, Graphite

Logitech MX Creative Console with 9 Customizable Keys, Graphite

As an affiliate, we earn on qualifying purchases.

From Contextuality to Larger Tests

Quantum contextuality describes how the outcome associated with a measurement can depend on which other compatible measurements are performed alongside it. In the grid game, fixed answers for every square cannot meet all the row and column rules: the row constraints require an even total number of minus signs, while the column constraints require an odd total. A classical strategy must consequently lose on at least one of the nine equally likely assignments.

The researchers used entangled qubits to produce answers that could satisfy the game more often than that classical limit. In the larger game, players received bits whose total was guaranteed to be even and had to return bits meeting a parity condition. The team used a shared GHZ state—a Greenberger-Horne-Zeilinger entangled state—with each player measuring a qubit based on their input.

The study also included the two-dimensional hidden linear function problem, which asks a quantum processor to return a bit string satisfying mathematical relationships encoded in an input table. The source report describes this test as a way to examine how much work a quantum computer can perform in a small number of steps. Its classical benchmark, however, remains less settled than the explicit game limits.

“The 105-qubit chip returned a valid answer more than half the time.”

— The study’s researchers, as summarized by The Quantum Insider

Limits of the 105-Qubit Comparison

The classical benchmark for the 105-qubit hidden linear function test is not proven: the estimate of how many steps a classical computer would require depends on a conjecture. The experiment’s more-than-half valid-answer rate is an observed result, but the size—and even the establishment—of a quantum advantage for that task remains open.

The report also does not establish that the game results translate into practical gains for commercial or scientific computing tasks. The measured contextuality score fell short of the ideal quantum value, and the team attributed that difference to experimental errors. Further work would be needed to assess how these tests perform across hardware generations and whether they reliably identify errors missed by other methods.

Further Hardware and Benchmark Tests

The study presents the games as tools researchers can use to test quantum behavior and potentially help locate hardware errors. The next steps are to repeat and extend these measurements, determine how performance changes as more qubits are involved, and assess whether contextuality tests add useful information to existing hardware checks.

For the hidden linear function experiment, a clearer classical comparison would be needed to establish the scale of any advantage. The reported work does not specify a date for follow-up experiments or a schedule for applying the proposed error-checking approach in production systems.

Key Questions

What did Google’s Willow chip demonstrate?

Researchers reported that Willow beat the classical winning-rate ceiling in two cooperative games where players could not communicate. In one, it won 98.3% of the time against a classical limit of 88.9%.

Does this prove Willow is faster than classical computers?

No. The games show performance beyond precise classical limits under their specific rules. They do not establish that Willow is faster or more useful than classical computers for general-purpose tasks.

What does the 105-qubit result show?

Willow returned a valid answer to a hidden linear function problem more than half the time in the reported experiment. The classical cost estimate relies on a conjecture, so the size of any advantage at that scale remains uncertain.

What do the 71-qubit and 45-qubit figures mean?

The game beat its classical limit with up to 71 qubits. Separately, the researchers verified genuine group entanglement among as many as 45 qubits; the figures describe different results.

Could these tests help find quantum hardware errors?

The researchers say contextuality tests may help identify errors that other checks miss. The report presents this as a possible application, not as a demonstrated replacement for existing error-detection methods.

Source: thequantuminsider.com

You May Also Like

How to Choose Smart Home Voice Assistants

Connect a voice assistant to your home Wi-Fi, add compatible devices, create routines, and check privacy and troubleshooting settings.

Onimusha Shows Why Capcom Is On A Winning Streak

The resurgence of Onimusha signals Capcom’s strong performance and renewed success in the gaming industry, reflecting a broader winning streak for the company.

Grok Bot Sends Tasks To Claude, Bringing AI Routing Into The Open

A Forkast.news headline says Grok Bot may route tasks to Claude, but the available material does not confirm the feature, its status or its scope.

How To Ask An AI Agent To Explain An Unfamiliar Project

A guide based on REA shows how to use coding agents to investigate one behavior at a time, check their evidence and avoid treating guesses as facts.