A new study suggests small quantum computers could process massive datasets more efficiently than exponentially larger classical systems by reducing memory requirements for key data tasks.
The researchers report that techniques such as quantum oracle sketching enable quantum systems to perform classification, dimension reduction and linear system solving using far fewer resources.
The findings are based on simulations and theoretical proofs, with practical impact dependent on future advances in quantum hardware, error correction and real-world validation.

In a lab on the western edge of Paris, where the River Seine flows wide and trams slide past glass-fronted buildings and blossoming cherry trees, a technician called Rémi makes some adjustments with a spanner.

The machine, a cascade of gold and silver-coloured cylinders descending through a cloud of wires, is a cryostat, its purpose: to create an ultra-cold environment.

At the bottom of the cylinder temperatures are down to minus 273 degrees Celsius. At that temperature thermal vibrations are minimised. Isolation from the outside world is complete.

In this cylinder is placed a small case, again of gold and silver colour, in which is inserted a chip.

This chip is another, tinier box inside of which takes place the phenomenon discovered by Albert Einstein and other physicists that seems to defy the mechanics of the world we live in: the quantum leap, when particles change energy levels in ways that are predictable, reproducible and apparently impossible.

This is the French quantum computer company Alice & Bob.

“Now we know they work, and in a few years we will have reliable quantum computers that we can hook up to High Performance Computers (HPCs) in data centres to exponentially increase their computing power,” says Peronnin.

“It’s not about being faster. It’s about being so dramatically faster that you change what is feasible. We will be able to solve problems that are absolutely intractable with classical computers,” he says.

Quantum computers struggle with a major flaw: their information vanishes unpredictably. Scientists have now created a new method that can measure this loss over 100 times faster than before. By tracking changes in near real time, researchers can finally see what’s going wrong inside these systems. This could be a big step toward making quantum computers stable and practical.

Insider Brief

World Quantum Day, observed on April 14 to reflect the Planck constant (4.14 × 10⁻¹⁵ electron volts-seconds), highlights the growing real-world impact of quantum technology as it transitions from research to deployment.
Quantum technologies use principles such as superposition and entanglement to enable new types of computation, sensing and communication, with potential applications across industries despite current technical limitations.
The technology presents both opportunities and risks, including potential threats to existing encryption systems, driving efforts in post-quantum security, workforce development and broader public engagement.
Image: Google’s Doodle today pays homage to World Quantum Day. Google, itself, is one of the quantum industry’s hyperscalers.

Global commercial leader with 21 systems sold to 13 customers to date – including 4 out of the top 10 supercomputing centres globally.
Industrial leader with 15 systems delivered (largest number publicly disclosed by selected quantum companies1), 30+ computers built, own chip factory and quantum data centre.
The transaction values IQM at a pre-money equity valuation of approximately USD 1.8 billion and makes IQM the first European quantum company to go public.
With the close of this transaction, IQM’s cash position expected to exceed USD 450 million.2
Significant business momentum, with at least USD 35 million3 2025 revenue (unaudited) and over USD 100 million bookings / visibility as of year-end 2025.
Strong commercial integrations with high-performance computing and enterprise platforms across the quantum/AI value chain such as NVIDIA, Hewlett Packard Enterprise, AWS, Toyo Corporation and Bechtle AG.
Technical successes, achieving greater than 99.9% fidelity for single-qubit and two-qubit gates and readouts in their processors, and on track to deliver broad commercialization with the release of its next generation system, Halocene.

Insider Brief

Classical computers process information using bits that exist as definite 0s or 1s and follow deterministic logical operations, while quantum computers use qubits that can exist in superposition, leverage entanglement, and exploit quantum interference to explore computational spaces that classical computers cannot efficiently navigate.
The relationship between quantum and classical computing is complementary rather than competitive, with quantum computers poised to serve as specialized co-processors for narrow problem classes like optimization, sampling, and quantum simulation rather than general-purpose replacements for classical systems.
Current quantum computers operate in the Noisy Intermediate-Scale Quantum (NISQ) era with error rates millions of times higher than classical processors, requiring fault-tolerant error correction and millions of physical qubits before they can consistently outperform classical supercomputers on practical applications.
The future computing landscape will likely feature hybrid architectures where classical computers handle most workloads – data processing, machine learning, business applications – while quantum processors tackle specific bottlenecks where quantum algorithms offer exponential or polynomial speedups.