August 11, 2026 -- Two recent scientific publications from the Fraunhofer Institute for Applied Solid State Physics IAF shed light on how the concept of quantum advantage can be assessed more precisely and realistically in the future. Both papers provide theoretical tools to make claims about quantum advantage more robust. Their content is complementary: one addresses quantum chemistry beyond idealized system models, while the other examines how algorithmic advantages can be reliably demonstrated through the scaling of problem sizes.

Quantum advantage refers to the point at which a quantum computer solves a clearly defined task faster or more efficiently than any classical computer—or makes it solvable in the first place. For many practical applications, this has not yet been demonstrated. Research therefore relies heavily on theoretical models and simulations to explore where and under what conditions such an advantage may realistically be achieved in the future.

Quantum simulation is considered a promising path toward genuine quantum advantage. However, many existing approaches in quantum chemistry rely on simplifying assumptions: they describe molecules as closed systems perfectly isolated from their environment, model only unitary dynamics, and focus on calculating ground states within the Born-Oppenheimer approximation. In nature, none of these assumptions fully hold.

This is precisely where the newly published review, "Beyond Unitary Quantum Simulation: Open-System Approaches for Quantum Chemistry Toward Quantum Advantage," comes in—a collaborative effort by experts from industry (HQS Quantum Simulations), academia (ETH Zurich), and applied research (Fraunhofer IAF). It challenges the common practice of approaching quantum chemistry primarily through idealized, closed systems and advocates for a fundamental shift in perspective.

Toward more robust quantum algorithms through open system dynamics

In reality, molecules and materials constantly interact with their environment: they release energy, relax, and often reach stable or thermal states precisely through this open dynamics. For quantum chemistry, solid-state physics, and materials science, such dissipative processes are not a marginal phenomenon but are often physically central. In practice, however, this idea has been implemented only in a limited form across much of the literature, since quantum algorithms are typically formulated for the Hamiltonian dynamics of closed systems.

Dissipation as a resource, not a disturbance

The review's central thesis is that dissipation and open system dynamics in quantum chemistry should no longer be viewed primarily as sources of disturbance. When applied in a controlled manner, they can become a genuine resource for quantum algorithms—for example, to prepare, stabilize, or sample from chemically relevant quantum states.

"The exciting question is not just whether quantum computers can outperform classical computers, but when, why, and under what conditions," says Dr. Florentin Reiter, co-author and head of the Quantum Systems business unit at Fraunhofer IAF. "For chemistry, this means we should not only consider idealized, closed systems but also the open dynamics that are ubiquitous in nature."

The new review is set within a broader research context, connecting work on open quantum systems, dissipative state preparation, and engineered dissipation with current questions surrounding fault-tolerant quantum algorithms, quantum machine learning, and QAOA. The overarching message is that quantum computers do not need to prove themselves in idealized models, but under realistic physical and algorithmic conditions.

QAOA: Demonstrating Quantum Advantage Through Scaling

A second publication from Fraunhofer IAF, authored by Vanessa Dehn, takes a different approach: algorithmic scaling. The paper, titled "Extrapolation method to optimize linear-ramp quantum approximate optimization algorithm parameters: Evaluation of runtime scaling," examines the optimization potential of the Quantum Approximate Optimization Algorithm (QAOA) for combinatorial problems, such as those found in finance, logistics, network planning, materials design, and machine learning.

The central question is not whether QAOA works on small examples, but how its computational cost scales as problem size increases. Only by demonstrating that a quantum algorithm remains more efficient than classical methods for large instances can one provide reliable evidence of genuine quantum advantage. Based on simulations, the study shows that for portfolio optimization problems within the examined problem size, scaling advantages over classical algorithms may be possible. An extrapolation-based methodology allows algorithm parameters to be transferred from small to large problem sizes, representing an important step toward practical applicability.

"Small-scale demonstrations alone are not enough," says Vanessa Dehn, author and specialist in quantum hardware simulation. "The crucial question is what happens as a problem grows larger. That is exactly where it becomes clear whether an approach can become relevant in the long term."

Both papers pursue the same overarching goal: advancing the concept of quantum advantage from a broad promise to a sober, precisely measurable concept. Earlier work on quantum machine learning complements this context: it has examined mathematically provable advantages and provided data-driven insights into when quantum models can particularly effectively capture practically relevant structures. Together, these publications paint a nuanced picture of how quantum computing can move from the stage of theoretical promise into the realm of concrete, verifiable application advantages.