For years, the trapped-ion quantum computer has been trapped itself—not by physics, but by optics.
Walk into any of the handful of labs pursuing this approach, and you'd find yourself navigating what looks like a photonics factory floor: hundreds of lasers, each aligned to within nanometers, each stabilized against the faintest vibration, each individually controlled. The setup worked brilliantly for a dozen qubits, maybe two dozen. But scaling to the thousands needed for practical quantum computing? That felt less like an engineering problem and more like trying to conduct an orchestra where every musician required their own soundproof room.
Then something shifted. The fix, when it arrived, came not from some exotic breakthrough in laser physics but from a technology so ordinary it powers your microwave oven.
MAGIC—short for Magnetic Gradient Induced Coupling—replaces those sprawling laser arrays with microwave fields and magnetic gradients. While the semiconductor industry has long worked with radio-frequency technologies, the specific application of static or oscillating magnetic field gradients combined with microwave control represents a quantum computing innovation. What sounded at first like an incremental refinement is starting to look transformative. Last October, IonQ announced it had hit 99.99% fidelity on two-qubit gates using electronic control on semiconductor-fabricated chips. That's the threshold most researchers agree puts fault-tolerant quantum computing within reach, not someday, but potentially within the next few years.
The timing is hardly coincidental. McKinsey's Quantum Technology Monitor, released in late April, reported quantum startups pulled in more than $12 billion during 2025—roughly six times higher than in 2024. For the first time, quantum-computing companies collectively cleared $1 billion in revenue. Money, it turns out, flows toward companies that appear to have solved—or at least credibly claim to be solving—the scalability problem that has bedeviled the field since its inception.
The Problem That Wouldn't Go Away
Trapped-ion systems have always had advantages their rivals envied. Because they manipulate individual atoms held in electromagnetic fields, they achieve pristine quantum states and full connectivity between qubits—no messy wiring, no crosstalk. The problem was never the qubits themselves. It was controlling them.
Traditional Raman gates use precisely tuned laser beams to drive quantum operations. At small scale, this works beautifully. But those same lasers introduce spontaneous-emission errors as systems grow, and the optical infrastructure required doesn't translate to anything resembling chip-scale manufacturing. You can't exactly stack thousands of individually aligned laser beams in a clean room and call it mass production.
MAGIC sidesteps the entire problem by using microwaves instead—the same radio-frequency signals that run cell towers. The theoretical roots stretch back more than two decades: spin-spin interactions in magnetic gradients were proposed in 2003, followed by on-chip gradient addressing in 2008 and microwave logic gates in 2011. By 2016, teams at Oxford and NIST had demonstrated near-field microwave gates in calcium ions with 99.7% fidelity. NIST published a comprehensive scheme for "high-fidelity laser-free universal control" in 2021, combining RF magnetic gradients with microwaves in ways that finally looked practical.
That academic lineage matters because it maps directly onto today's commercial landscape. Oxford Ionics, founded by Dr. Chris Ballance and Dr. Tom Harty—both veterans of those early Oxford and NIST experiments—built its entire company around the premise of "electronics instead of lasers." In Germany, eleQtron spun out of the University of Siegen's quantum optics group, which spent over a decade refining MAGIC techniques. Universal Quantum, emerging from the University of Sussex, has been advocating for microwave-based control in its modular architectures since its founding.
Recent results suggest the technology is maturing faster than many expected. In February 2024, researchers demonstrated cryogenic surface-trap microwave gates at roughly 1% error in 154 microseconds and 0.5% error in 331 microseconds—results published in Physical Review A that July. Oxford Ionics presented 100-microsecond two-qubit gates at the DAMOP conference that same year. By June 2025, a Siegen group published work in Physical Review X on "Fast, Robust, and Laser-Free Universal Entangling Gates," pushing both speed and robustness forward.
The cadence feels different now. Less academic, more urgent.
Why Now? Three Converging Forces
The shift from lasers to microwaves isn't happening in a vacuum. It's being driven by performance limits, manufacturing realities, and—perhaps more than the founders expected—investor appetite.
Start with performance. MAGIC gates avoid the spontaneous-emission errors inherent in Raman transitions. They tolerate higher ion temperatures, which eases cryogenic requirements and associated costs. A 2017 analysis in New Journal of Physics laid out the engineering rationale: microwave control integrates naturally with surface-electrode traps and RF electronics, technologies the semiconductor industry already knows how to manufacture at scale. More recent work from groups using superconducting surface-trap chips and integrated microwave resonators—published in 2024 and 2025—demonstrates that standard semiconductor processes can fabricate the necessary components.
The German Aerospace Center's Quantum Computing Initiative published a microwave frontend design in March 2025 targeting MAGIC-based processors. It's a component-level enabler that achieved idle gate fidelities up to 99.99% at low output power. These aren't laboratory curiosities anymore; they're engineering solutions designed for reproducibility, for supply chains, for procurement departments.
Manufacturing is where things get interesting. Infineon, Europe's semiconductor heavyweight, has been partnering with both eleQtron and Oxford Ionics to industrialize trapped-ion quantum processing units, leveraging its expertise in semiconductor fabrication. Infineon's April update cited involvement in three EU quantum pilot lines—SUPREME, CHAMP-ION, and SPINS—where the stated goal is 24/7 user access and modular scale-out. The company's promotional materials frame trapped ions as a "lab to fab" pathway, leveraging existing chip fabrication infrastructure worth billions.
That resonates with investors in ways laser-based approaches never quite did. Forrester's "State of Quantum Computing, 2026" report, published in March, asserts the industry has entered a "fault-tolerant foundation era" where metrics are shifting from raw physical qubit counts—always a bit of a vanity metric—to error-corrected logical qubits that can actually do useful work. Forrester's analysis suggests a potential impact around 2030, when quantum computers could potentially break current encryption. Enterprise pilots in pharma, financial services, and materials science are multiplying. CTOs want systems that can scale without exponential cost or Rube Goldberg complexity.
Then there's the capital momentum. Quantinuum, the Honeywell spin-off pursuing trapped-ion computing (albeit still primarily with lasers), raised $600 million in September 2025 at a $10 billion pre-money valuation. That deal signaled investor confidence in trapped-ion roadmaps broadly—and by extension, in any technology that makes those roadmaps more credible.
Three Companies, Three Bets
IonQ made the most audacious move. In June 2025, it announced a $1.075 billion acquisition of Oxford Ionics. The deal closed in September—with the final purchase consideration reported in IonQ's 2025 Annual Report, filed this May, at approximately $1.59 billion, the difference reflecting share price fluctuations at closing. Oxford Ionics brought chip-integrated microwave antenna technology and a team that had essentially invented modern near-field control techniques.
Four months later, on October 21, IonQ announced its 99.99% fidelity milestone, explicitly crediting what it now brands "Electronic Qubit Control"—the company's label for microwave-driven gates on semiconductor chips. On April 22, IonQ published what it called a "definitive technical report" outlining a scaling path to more than 10,000 physical qubits and detailing its error-correction trajectory. CEO Niccolo de Masi framed the blueprint as proof that electronic control isn't just competitive—it's superior for modularity and scale.
IonQ's customer list reads like a who's-who of enterprise tech: AWS, AstraZeneca, NVIDIA. It signed agreements with South Korea's KISTI for a Tempo system delivery and holds Air Force Research Laboratory contracts valued at $25.5 million. Whether all that translates into sustainable revenue remains to be seen, but the optics—or rather, the lack of optics—are compelling.
eleQtron, meanwhile, operates as the Siegen group's commercial vehicle. On May 5, it closed a €57 million Series A, and the University of Siegen announced the company now employs more than 100 people—a startup hitting manufacturing scale. eleQtron's proprietary MAGIC approach, refined over years of academic work, targets scalable processors with radically reduced optical complexity. The Infineon partnership gives it a semiconductor manufacturing pathway that few quantum startups can claim. Its involvement in German and EU quantum programs—ATIQ, MILLENION, and DLR QCI projects running through this year—positions it as something of a national champion in Europe's race to industrialize quantum computing.
Universal Quantum's CEO, Sebastian Weidt, has been more publicly critical than most of his peers. In a May 2025 blog post and an April op-ed, Weidt called for "industry-wide focus on scalability and honest timelines," arguing that microwave-based control and modular QCCD (quantum charge-coupled device) architectures are the only credible paths forward. It's the kind of blunt talk you don't often hear in a sector prone to breathless optimism.
Universal Quantum holds a €67 million contract from Germany's DLR to build scalable trapped-ion systems. Its public materials emphasize UQLogic (microwave control) and UQConnect (ion-shuttling interconnects). The company's founders—Weidt and Prof. Winfried Hensinger—emerged from Sussex's Ion Quantum Technology group, which has been researching microwave gates since the mid-2010s. They've been at this long enough to know what doesn't work.
Not every trapped-ion player is betting exclusively on MAGIC, of course. Quantinuum launched its System Model H2 "Helios" on November 5 last year, with enterprise customers including Amgen, BMW, JPMorgan Chase, and SoftBank. Quantinuum's systems still rely primarily on optical control but have achieved record quantum volumes—33,554,432 in September 2025, according to company statements. Alpine Quantum Technologies added its 12-qubit calcium-ion system to Amazon Braket in November, expanding cloud access. These companies prove laser-based approaches remain viable, at least for now.
The Road Ahead—and the Roadblocks
The question isn't whether MAGIC works anymore. The question is whether it scales faster than competing modalities can optimize their own architectures.
Superconducting qubits (IBM, Google) and neutral atoms (QuEra, Pasqal) have their own scaling narratives, each with advantages and limitations. Photonic approaches (PsiQuantum, Xanadu) promise room-temperature operation, if they can solve the integration challenges. But trapped ions still hold fidelity and connectivity advantages that are hard to dismiss. McKinsey's update this spring revised its 2035 internal market forecast upward, citing accelerating deployment readiness. QED-C's "State of the Global Quantum Industry 2026," published in mid-April, provides segment-by-segment market forecasts and workforce trends—all suggesting the industry has moved past speculative proof-of-concept into something resembling commercial execution.
The regulatory environment is tightening in ways that benefit the entire sector. NIST finalized post-quantum cryptography standards in August 2024 (FIPS 203, 204, and 205), and a White House Office of Management and Budget memo from November 2022 directs federal agencies to inventory vulnerable systems and plan migrations. That's creating enterprise demand for quantum-safe security programs—and indirectly, for quantum technologies more broadly.
In the U.S., the National Quantum Initiative Reauthorization Act is moving through Congress (S.3597, introduced in February, with markup in late April). If passed, it would extend and expand programs at NIST, NSF, and DOE, and authorize quantum communications corridors. The UK committed £2.5 billion over a ten-year period starting in 2024 in its National Quantum Strategy, with missions targeting large-scale error correction by 2032. The EU is working toward a "Quantum Act" proposal this year to coordinate infrastructure and skills across member states.
These programs matter because they de-risk private investment and create what industry insiders call "pull-through demand." National labs and defense agencies—such as AFRL's contracts with IonQ and the UK's National Quantum Computing Centre procurement of Oxford Ionics systems—serve as anchor customers while the commercial market finds its footing.
But the technical challenges ahead aren't trivial. Trap engineering—managing electric-field noise, ion heating, integrating on-chip features—remains a critical constraint, as a 2021 Nature Reviews Materials analysis laid out in detail. Control electronics and RF frontends must mature from lab prototypes to production-ready modules, a transition that has killed more than one promising technology. Active research continues on marrying QCCD ion-shuttling with in-chip microwave control, as discussed in a May seminar at Duke's Quantum Center. Speed-versus-fidelity trade-offs persist; even as microwave gates approach 100-microsecond operation, optical Raman gates can be faster in some regimes.
Then there's the question of honesty. Weidt's repeated calls for "honest timelines" reflect a growing impatience with qubit-count hype. Forrester's assertion that metrics are shifting to error-corrected qubits—not raw physical qubits that can't do much beyond generate press releases—signals a reframe. IonQ's decision to publish a detailed fault-tolerant quantum computing blueprint in April is unusual in an industry where vagueness has historically been a feature, not a bug. These are signs of a sector transitioning from research theater to engineering discipline, though whether that transition completes remains an open question.
MAGIC won't be the only answer, and anyone claiming otherwise is selling something. Sandia and Quantum Scientific Alliances revealed in June 2025 that they've tested trap architectures capable of storing up to 200 ions—a brute-force scaling approach that could complement microwave control rather than replace it. MIT announced in January that its integrated photonics advances could enable chip-based cooling and readout, addressing another piece of the puzzle. Even MAGIC systems still need lasers for ion cooling and state readout; the "laser-free" label is shorthand for "laser-minimized gates," which is less catchy but more accurate.
What's becoming clear, though, is that the laser problem—once treated as an existential constraint on trapped-ion computing—is yielding to engineering ingenuity in ways few predicted even five years ago. The companies raising hundreds of millions, the semiconductor giants building pilot lines, the researchers publishing gate fidelities at or near fault-tolerant thresholds—they're all betting that microwave control isn't just viable. They think it's inevitable.
Whether that bet pays off by 2030, or 2035, or later still, the trajectory has unmistakably shifted. Scalability, for trapped-ion quantum computing, is no longer a distant aspiration relegated to conference panels and white papers. It's a procurement specification, a line item in government contracts, a manufacturing challenge with timelines and milestones.
Which means, perhaps, that the forest of lasers that defined quantum computing's first era is finally being cleared.
