Level 2 - Elementary

Scientists at the Flatiron Institute in New York made a major discovery in May 2026. They created a new computer algorithm that can solve problems once thought impossible for regular computers.

A company called D-Wave claimed in March 2025 that only its special quantum computer could solve certain difficult physics problems. The Flatiron team proved this wrong. Their classical algorithm, running on a laptop or desktop computer, achieved the same results.

The research was published in the journal Science on May 21, 2026. It is a major moment for the field of quantum computing. Scientists now need to rethink which problems truly require a quantum computer to solve.

algorithm

a computer program or set of rules for solving a specific problem step by step

quantum computing

a type of computing that uses quantum physics to process information in new ways

classical computing

regular computing using standard computer chips and binary code

tensor network

a mathematical tool used to represent complex quantum states more efficiently

simulation

using a computer to model or reproduce a real-world system or phenomenon

benchmark

a standard test used to compare the performance of two different systems

collaboration

a group of people or institutions working together toward a shared goal

claim

a statement that something is true, usually one that needs to be verified or proven

Level 3 - Intermediate

A team of physicists at the Flatiron Institute, in collaboration with Boston University, made headlines in May 2026 after publishing a study in the journal Science that directly challenged a celebrated quantum supremacy claim. The research demonstrated that a cleverly designed classical algorithm could simulate complex three-dimensional quantum dynamics with accuracy equal to D-Wave's 5,000-qubit Advantage2 machine.

D-Wave Systems, the Canadian quantum computing pioneer, had announced in March 2025 that its Advantage2 processor solved a physics problem beyond the reach of any classical computer, a milestone known as quantum supremacy. The Flatiron team decided to test this claim using a technique called tensor networks, a mathematical method for compressing the description of quantum states into a form manageable on ordinary hardware.

By combining tensor network algorithms with a mathematical routing technique called belief propagation, the Flatiron researchers achieved matching accuracy on standard workstations and even ordinary laptops. The finding, published by the Simons Foundation's Center for Computational Quantum Physics, reshapes the debate over what quantum computers truly offer over their classical counterparts and opens new avenues for simulating quantum materials on conventional hardware.

quantum supremacy

the ability of a quantum computer to solve a problem that a classical computer cannot solve in a practical timeframe

tensor network

a mathematical framework representing quantum states as interconnected arrays of numbers

belief propagation

a message-passing algorithm that routes information across a network to minimize error

qubit

the basic unit of quantum information, analogous to a classical computer bit

Hilbert space

the mathematical framework describing all possible states of a quantum system

combinatorial optimization

finding the best solution from among a very large number of possible combinations

benchmark problem

a specific test case used to measure and compare the performance of different systems

frustrated magnetic system

a physical system where competing interactions prevent particles from settling into a simple low-energy state

Level 4 - Advanced

A landmark study published in Science on May 21, 2026 by researchers at the Flatiron Institute's Center for Computational Quantum Physics and Boston University administered a significant setback to the quantum supremacy narrative, demonstrating that a 3D tensor-network classical algorithm can reproduce D-Wave's Advantage2 results on a physics problem the company declared beyond classical reach in March 2025. The collaboration, anchored at the Simons Foundation, directly targets D-Wave's 'beyond-classical' milestone and argues that the company's prior benchmarking methodology failed to adequately explore the classical-algorithm solution space.

The Flatiron team repurposed and optimized two mature but underutilized computational techniques: three-dimensional tensor networks, which represent quantum states as networks of contracted tensors enabling exponential compression of the Hilbert space description; and belief propagation, a message-passing algorithm borrowed from statistical physics that routes information across the network to minimize contraction error. The resulting algorithm runs on commercial workstations and achieves state-of-the-art accuracy on the frustrated magnetic problem D-Wave had targeted, specifically the sampling of low-energy configurations of a three-dimensional Ising Hamiltonian defined on the Pegasus graph topology.

The implications extend well beyond the immediate dispute. Quantum annealing, D-Wave's hardware paradigm, is designed to find low-energy states of combinatorial optimization problems; if classical algorithms can match its accuracy on benchmark instances, the commercial case for near-term quantum advantage in optimization narrows considerably. D-Wave's CEO is preparing a response, and the broader community expects a counter-paper specifically addressing whether the Flatiron algorithm scales to the larger 7,000-plus-qubit Advantage3 system currently in field trials.

More broadly, the episode illustrates the methodological challenge that confronts every quantum computing supremacy claim: supremacy is a relative statement that depends entirely on the reference classical algorithm, and improvements in tensor-network and Monte Carlo methods have repeatedly closed gaps that seemed definitive. The field is left with a refined question: not 'can quantum computers do what classical ones cannot' but 'at which problem sizes, error rates, and physical-system descriptions does the quantum approach irreversibly outperform the best classical alternative?' That narrower question will likely define the research agenda for the next decade.

Hamiltonian

a mathematical operator representing the total energy of a quantum system

Hilbert space

the complete mathematical space describing all possible quantum states of a physical system

tensor contraction

the mathematical process of multiplying and summing tensors together to reduce the complexity of a network

quantum annealing

a computational method that slowly lowers energy to find the lowest-energy solution to an optimization problem

Pegasus graph

the qubit connectivity topology used in D-Wave's Advantage family of quantum processors

near-term quantum advantage

a computational benefit from current quantum hardware before full error correction is achieved

Monte Carlo method

a computational technique that uses random sampling to approximate numerical solutions

benchmarking methodology

the systematic approach and criteria used to measure and compare computational performance