Quantum Medicine: Designing the Drugs Today’s Computers Cannot See

The next breakthrough medicine may begin with a calculation today’s computers cannot finish.

Somewhere today, researchers are asking a quantum processor how an mRNA strand might fold, which peptide could train the immune system to attack a tumor, or how a potential drug interacts with a protein complex containing more than 12,000 atoms. These calculations become extraordinarily difficult because every additional atom, bond, and possible molecular shape creates another layer of interactions. Eventually, even the world’s most powerful conventional computers must simplify the problem, make approximations, or politely request several more centuries.

A traditional computer processes information using bits that are either a zero or a one. A quantum computer uses qubits, which can represent combinations of states through a property called superposition. Qubits can also become linked through entanglement, allowing quantum algorithms to represent and manipulate complex relationships in ways ordinary bits cannot. The physics becomes strange rather quickly, which is generally a reliable sign that someone nearby is about to ask for a larger research grant.

For a broader explanation of the hardware, I previously wrote about Google’s progress in “Meet Willow: Google’s New Chip That’s 13,000× Faster and Still Can’t Help You Log Into Outlook.” Willow showed how quickly quantum systems are improving at specialized calculations. Medicine may eventually become one of their most consequential applications.

Quantum-enabled drug discovery remains in its earliest industrial phase. Today’s most promising projects combine quantum processors with classical supercomputers and artificial intelligence, assigning each system the portion of the problem it handles best. The machines are specialized, the experiments remain narrow, and the first useful results are beginning to arrive.

Where Quantum Medicine Is Taking Shape

• IBM and Moderna—predicting how mRNA folds. The shape of an mRNA strand affects its stability, how efficiently it produces proteins, and how it behaves inside the body. IBM and Moderna used IBM quantum processors to examine mRNA structure-prediction problems involving as many as 80 qubits, producing minimum-energy structures that matched results from a leading classical solver. More recent hybrid work has extended these methods to longer sequences, creating a possible path toward more efficient design of future mRNA vaccines and therapies.
• Cleveland Clinic, IBM and RIKEN—modeling drug interactions across more than 12,000 atoms. Researchers divided two large protein–ligand systems into smaller fragments, processed portions of the electronic calculations on IBM quantum computers, and reassembled the results using two major classical supercomputers. The larger complex contained 12,635 atoms, making it the largest biologically meaningful molecular system reported at the time using quantum hardware. The achievement provides a scalable framework for studying how drugs bind to proteins while giving the machines involved a workload considerably more demanding than updating a spreadsheet. 
• Q-CHIPP—identifying cancer targets the immune system may recognize. A tumor can produce thousands of abnormal peptides, while only a small fraction will provoke a useful immune response. Cleveland Clinic and IBM developed a quantum machine-learning framework that combines predictions of whether a peptide will be displayed by a cancer cell and whether the immune system is likely to recognize it. In a 46-qubit hardware experiment, the model improved classification accuracy by six percentage points with fewer training samples than comparable classical approaches and identified relevant peptides missed by benchmark models. That could eventually help researchers design more personalized cancer vaccines and immunotherapies.
• Technical University of Denmark and ORCA Computing—designing immune-targeting peptides. Researchers combined generative AI with a photonic quantum processor to generate peptides intended to bind to specific immune-system proteins. The quantum-guided model reportedly produced stronger results than its classical counterpart, particularly when training data was scarce. That matters because biological datasets are often small, expensive, and heavily concentrated in certain populations; unfortunately, the human immune system has never been especially considerate about providing researchers with a clean Excel file.
• Algorithmiq, Cleveland Clinic and IBM—studying light-activated cancer drugs. Photodynamic therapy uses molecules that remain inactive until exposed to a particular wavelength of light, after which they can damage cancer cells. Modeling these excited molecular states is exceptionally difficult for classical methods, especially when transition metals and complex electronic behavior are involved. Algorithmiq developed a hybrid workflow that used real quantum hardware alongside classical calculations to study these reactions, reporting comparable or improved accuracy with lower classical complexity in parts of the simulation.
• Quantum-assisted drug design beyond the major collaborations. Researchers have also used hybrid quantum-classical generative models to propose new inhibitors for KRAS, an important cancer target, and then synthesized and tested the resulting compounds. Two candidates demonstrated measurable engagement with KRAS proteins, offering an early example in which a quantum-influenced design workflow produced experimentally confirmed biological hits. These compounds remain far from approved medicines, though the research shows that quantum systems are beginning to participate in the actual search for viable molecules rather than remaining confined to theoretical demonstrations. 

The Third Machine in the Medical Factory

This article completes a progression that began with “The Next Medical Revolution Has Already Entered the Waiting Room.” That first piece examined how AI is already helping researchers discover drugs, interpret medical images, identify disease earlier, and improve clinical trials. Artificial intelligence brings pattern recognition and prediction to medical problems that previously depended on slower, more sequential human analysis.

The second article, “The Robot Will See You Now: Scalpels, Robots and the Operating Room of Tomorrow”, moved the revolution into the physical world. Robotic platforms are giving surgeons steadier instruments, smaller movements, improved imaging, and increasingly intelligent assistance during procedures. Research systems are even beginning to perform defined surgical steps autonomously under controlled conditions.

Quantum computing adds a third capability: the potential to model forms of molecular complexity that eventually overwhelm conventional computing. AI can search for patterns among enormous amounts of data. Quantum systems may help calculate the underlying chemistry with greater precision. Robots could eventually execute the resulting treatments with extraordinary physical accuracy. Together, these technologies begin to form a connected medical system stretching from the first molecular hypothesis to the treatment delivered inside the body.

Consider what that could mean over time. A quantum system might model how thousands of potential molecules interact with a difficult disease target. AI could narrow those possibilities, predict toxicity, and identify the patients most likely to respond. Automated laboratories could manufacture and test candidates, while robotic systems perform increasingly precise procedures guided by real-time data. The full process could become faster, more adaptive, and far less dependent on years of expensive trial and error.

Considerable work remains. Quantum processors still suffer from noise, limited qubit counts, short operating windows, and the engineering challenge of error correction. Most medical problems will continue to rely heavily on classical computers and AI for many years. New methods must also survive laboratory validation, clinical trials, regulatory review, hospital integration, and reimbursement—a sequence capable of making quantum physics look refreshingly straightforward.

For investors, the opportunity reaches across quantum hardware, control systems, software, hybrid cloud infrastructure, semiconductors, pharmaceutical companies, biotechnology firms, and specialized algorithm developers. Durable winners will need to convert technical milestones into commercially useful tools, a step that tends to receive less attention than the original scientific announcement and considerably more scrutiny from the income statement. Patience, diversification, and a healthy respect for valuation remain essential.

As always, please call or email us if you would like to discuss how advances across AI, robotics, and quantum computing may intersect with your financial plan or how we are positioning portfolios around the companies enabling this medical transformation. Quantum systems can model thousands of interacting atoms and operate at temperatures near absolute zero, though they still cannot explain why the printer says it is offline while sitting three feet away. Human assistance therefore remains useful.

Your capital, our expertise, a bespoke creation.