Quantum Computing: The Compute Power Behind “Curing Cancer”

A few weeks ago, my boss Cassidy posted a video about her feelings on AI. She called it “An attempt at a balanced perspective on AI” and described the process as repeatedly “crashing out” while working through her thoughts.

I watched it. Then I left a comment: “Let’s get AI to cure cancer first, then throw it in the ocean.”

I’m a breast cancer survivor. That experience rewrites your priorities. When people ask what I want from AI, I don’t say better autocomplete. I say cure cancer. Then throw it in the ocean. But cure cancer first.

That comment stuck with me. I started wondering what it would actually take. Not the hype. Not the TED talks. The actual compute. And that question led me somewhere unexpected: quantum computing.

What Quantum Computing Actually Is

Let’s clear something up. Quantum computers are not faster regular computers. They’re fundamentally different machines that solve fundamentally different problems.

Classical computers use bits. Each bit is either 0 or 1. You know this. It’s how every computer you’ve ever used works.

Quantum computers use qubits. A qubit can be 0, 1, or both at the same time through a property called superposition. When qubits interact through entanglement, they can represent and process vastly more information than the same number of classical bits. The math gets weird fast. But the intuition matters: quantum computers don’t try every possibility sequentially. They explore probability spaces in ways that classical computers physically cannot replicate.

This isn’t theoretical anymore. Google’s Willow chip demonstrated a 13,000x speedup over the world’s fastest supercomputer in October 2025 for a specific quantum algorithm. That’s not a rounding error. That’s a different category of computation.

Where We Are in 2026

The quantum computing narrative has shifted from “maybe someday” to “which modality wins.”

Caltech’s 6,100-Qubit Array

Caltech built a neutral-atom array with 6,100 qubits in September 2025, the largest controlled quantum system ever assembled. The previous record was around a thousand. These qubits maintained coherence for 13 seconds with 99.98% fidelity. Coherence time matters because it determines how much useful computation you can do before quantum effects collapse.

IBM’s Nighthawk Processor

Running at 120 qubits, achieved a 10x speedup in quantum error correction decoding. Error correction is the bottleneck. Quantum states are fragile and every interaction with the environment introduces noise. IBM’s progress means we’re getting better at keeping quantum information intact long enough to compute with it.

Quantinuum’s H2 Processor

Became the first quantum computer to reach Microsoft’s Level 2 Resilient phase. They produced logical qubits with error rates 800x lower than physical rates using just 30 physical qubits to create four logical qubits.

The Industry Crossed a Billion Dollars in Revenue

IonQ, the largest publicly traded pure-play quantum company, hit $39.9M in Q3 2025 revenue, up 221% year-over-year. Governments have committed over $40 billion in national quantum strategies. These are not vaporware numbers.

Why Drug Discovery Changes Everything

Here’s where cancer comes in.

The pharmaceutical industry has a problem: most AI-designed drug molecules look promising on computers but are nearly impossible to synthesize in labs. This is called downstream attrition. You find a candidate that should work, spend millions developing it, then discover it can’t actually be manufactured.

Classical computers struggle with molecular simulation because the math of quantum mechanics scales exponentially with the number of electrons. Simulating a single caffeine molecule accurately requires classical computing resources that push against the limits of what’s physically possible.

Quantum computers are naturally suited for this. They don’t simulate quantum mechanics. They are quantum mechanics.

In January 2026, PolarisQB released a head-to-head study showing their quantum annealing platform running on D-Wave hardware outperformed classical generative AI for drug discovery. Classical AI took 40 hours to suggest molecules. The quantum system took 30 minutes. More importantly, the quantum-generated leads were significantly easier to synthesize in the lab.

In February 2026, Telefónica, Vithas hospitals, and Francisco de Vitoria University launched a quantum computing project specifically for cancer drug design. They’re targeting the BRAF V600E mutation, an altered protein that drives uncontrolled cancer cell growth. The goal is to use quantum algorithms to generate molecules that inhibit this protein. This is not a whitepaper. It’s happening now.

The Compute Requirements

I asked: what would it actually take to cure cancer with compute?

The honest answer is we don’t know yet. But we’re getting real data points.

The number of possible drug-like molecules exceeds 10^60. That’s more than atoms in the observable universe. Classical computers cannot search this space exhaustively, so they use heuristics, approximations, and machine learning to guess at promising regions. Quantum computers approach this differently. For specific optimization problems, and drug discovery is fundamentally an optimization problem, they can explore solution spaces in ways classical algorithms cannot match.

Researchers at Caltech published a quantum computing framework in February 2026 specifically for multi-stage drug discovery, covering allosteric site identification, protein-peptide docking, and molecular dynamics. It encodes three quantum algorithms that classical methods struggle to approximate.

The scale needed for practical advantage keeps shrinking. Google’s demonstration used their 105-qubit Willow chip. IBM’s roadmap targets 200 logical qubits by 2028. Logical qubits are error-corrected qubits that can run reliable computations. Each logical qubit requires many physical qubits to maintain, but the ratios are improving fast.

The Hardware Reality

Different companies are betting on different qubit technologies, and none have clearly won.

Superconducting Qubits (IBM, Google, Rigetti) Offer fast gate operations but require millikelvin temperatures. IBM operates the largest cloud-accessible quantum fleet and Google achieved the landmark quantum advantage demonstration with this approach.

Trapped Ions (IonQ, Quantinuum) Offer higher fidelity and fully connected qubits at the cost of slower operations. IonQ holds the world record at 99.99% two-qubit gate fidelity and Quantinuum’s H2 processor leads in quantum volume.

Neutral Atoms (QuEra, Pasqal, Atom Computing) Can pack thousands of qubits in arrays with longer coherence times. This is the technology behind Caltech’s 6,100-qubit breakthrough.

Photonic Computing (PsiQuantum, Xanadu) Has room temperature operation potential but harder gate reliability. PsiQuantum raised over $2 billion and is building datacenter-scale quantum compute centers.

All of these are viable for different applications. The race is still open.

A New Research Agenda

I started this with a frustrated comment about AI hype. I ended it with a reading list of quantum physics papers and a new appreciation for how much work happens between the hype cycles.

The quantum computing industry has moved from “trust us, it’s coming” to “here are the benchmarks.” The applications have narrowed from “everything” to specific domains where quantum mechanics gives inherent advantages. Drug discovery and materials science lead that list.

Cancer isn’t one disease. It’s thousands of molecular malfunctions, each requiring an understanding of specific protein interactions, cellular pathways, and drug binding mechanisms. Classical compute hits walls. Quantum compute might help us climb over them.

I’m not saying quantum computers will cure cancer. I’m saying the people who might cure cancer are starting to use them. That’s not AI hype. That’s a research direction worth following.

Andrea Griffiths is a Senior Developer Advocate at GitHub and a breast cancer survivor. She writes Main Branch, a newsletter about developer fundamentals.

Quantum Computing: The Compute Power Behind 'Curing Cancer'