Google Including Neutral Atom Qubit Approach
BigTech is charting their Quantum computing future more clearly in 2026. Many roadmaps point to 2030 as key year. However this full-stack dual qubit modality approach is promising.
It’s March, 2026 and Google Quantum’s roadmap is getting a lot more tangible. Google Quantum AI is broadening its quantum computing roadmap by introducing a neutral atom quantum computing programme alongside its established superconducting qubit research. This is of course validation for Neutral Atom approach to qubit development.
As you know in a nutshell, The Neutral Atom approach is a method of quantum computing that uses uncharged atoms—typically alkali metals like Rubidium or Cesium—as qubits. These atoms are suspended in a vacuum and manipulated using highly focused laser beams.
The elegance of this approach is that neutral atoms are not held by magnetic or electric fields, but rather by the "pressure" of light.
Which sounds almost poetic to me.
While Google’s Quantum unit is I think guilty of a lot of PR down the years, this is an interesting development signals to me that BigTech is taking Quantum computing and its future more seriously.
When Google Quantum acquired Atlantic Quantum back in October, 2025 I was a bit surprised.
Additionally Google’s introducing a 2029 timeline to secure the quantum era with post-quantum cryptography (PQC) migration. Read their official blog on this too. Google Quantum appears to see superconducting technology as the main path forward with Neutral Atom Qubit approach as more “complementary”.
Why add neutral atoms now?
Google is pursuing a dual-modality strategy because the two approaches have complementary strengths that address different scaling challenges, for instance:
Superconducting qubits (Google’s core platform):
Excel in the “time dimension” — fast gate/measurement cycles (~1 microsecond).
Have demonstrated millions of gate/measurement cycles in deep circuits.
Next challenge: scaling architectures to tens of thousands of qubits.
Neutral atom qubits (new addition):
Excel in the “space dimension” — already scaled to arrays of ~10,000 qubits.
Offer flexible any-to-any connectivity (no fixed wiring constraints), which enables more efficient algorithms and low-overhead error-correcting codes.
Trade-off: slower cycle times (~milliseconds).
Next challenge: demonstrating deep circuits with many sequential operations.
Quantum companies, startups and labs are notorious in making ambitious road-maps that don’t always turn into tangible Quantum breakthroughs and developments overall have been slow and shifting.
Quantum appears to be a bit of a sideshow as many BigTech companies including Nvidia, Amazon, Microsoft and Google. From their blog it’s clear that Google plans to develop the neutral atom platform through advances in error correction, simulation and hardware, while continuing its superconducting roadmap toward commercial systems by the end of the decade.
With Infleqtion going public, INFQ 0.00%↑, and Google giving the Neutral Atom approach a bit more mainstream credibility it’s fascinating to watch.
The Big Three in Neutral Atom Approach
QuEra
Pasqual
Infleqtion
It’s safe to say that these are the big three, more or less in the right order. Important to note that Pasqual is indeed also rushing to go public via a SPAC. The company will combine with Bleichroeder Acquisition Corp. II, a vehicle set up by Michel Combes and Andrew Gundlach, at a $2 billion pre-money valuation. While getting about $200 million in the process. I don’t believe I’ve covered the French team before - Cofounded in 2019 by Nobel prize winner Alain Aspect, Pascal builds and operates neutral-atom quantum processors. It has more than 275 employees and raised over $200 million from investors.
QuEra on the other hand, appears to have a pretty high ceiling with some very solid research. Based in Boston, QuEra is widely considered the technical leader in "analog" quantum simulation, though they are pivotally moving toward fault-tolerant digital systems.
Google Quantum’s dual approach to building a full-stack approach to scaling Qubits could be more important than it seems.
The Neutral Atom Approach Continued:
To create a quantum processor, scientists use Optical Tweezers. These are tightly focused laser beams that can grab individual atoms and move them into precise 2D or 3D geometries.
Qubit State: The quantum information is usually stored in the atom’s internal energy levels (hyperfine states) or by exciting the atom to a Rydberg state.
The Rydberg Blockade: This is the “secret sauce” for logic gates. When an atom is excited to a high energy level (a Rydberg state), its electron cloud expands significantly. This creates an electric field that prevents nearby atoms from being excited to that same state. This mutual influence allows two atoms to perform a “conditional” operation, forming a CNOT gate.
Google has launched a dedicated neutral atom hardware effort with three core pillars:
Quantum Error Correction — Adapting codes to the unique connectivity of neutral atom arrays for low space- and time-overhead fault-tolerant architectures.
Modeling and Simulation — Leveraging Google’s world-class compute resources for hardware architecture simulation, error budget optimization, and component targeting.
Experimental Hardware Development — Building the capabilities to control and manipulate atomic qubits at application scale with fault-tolerant performance.
Google’s Neutral Atom Team
The new team is based in Boulder, Colorado (a major hub for atomic, molecular, and optical physics) and is led by Dr. Adam Kaufman, a JILA Fellow and CU Boulder faculty member who is joining Google while maintaining his academic ties.
Google’s Space-Time Experiment in Quantum
Google frames this expansion as a trade-off between “Space” and “Time.” By pursuing both, they aim to cover the weaknesses of one with the strengths of the other:
Superconducting (The “Time” Leader): These qubits excel at circuit depth. They have extremely fast cycle times (microseconds) and have already demonstrated complex operations with millions of gate cycles. However, scaling them to the tens of thousands of physical qubits needed for error correction is a massive engineering hurdle.
Neutral Atoms (The “Space” Leader): These excel at qubit count. They have already scaled to arrays of ~10,000 qubits using “optical tweezers” (lasers) to hold individual atoms. While their cycle times are slower (milliseconds), they offer flexible, “any-to-any” connectivity, which makes error-correcting codes much more efficient.
This dual strategy might be mirrored by other BigTech teams I suspect as a new formula or modality in and of itself. Sort of like a full-stack approach to Quantum modalities and scaling.
Kaufman is a notable for his pioneering work in Atomic, Molecular, and Optical (AMO) physics. On this University of Colorado Physics page is reads:
“My research focuses on how to apply the tools of atomic, molecular, and optical physics to the microscopic investigation of quantum mechanics. I am interested in understanding and revealing the role of entanglement in complex quantum systems, both from a fundamental standpoint as well as for the purpose of understanding relevant condensed-matter models. I also investigate how to push the limits on our ability to retain quantum coherence while building up increasingly complex quantum states. To pursue these goals, we rely on microscopy, precision spectroscopy, and ultracold atom techniques.”
Here is a list of his publications.
The Boulder Colorado area is already a key hub in Quantum computing research and talent. Dr. Kaufman is widely known for his expertise in manipulating individual atoms using optical tweezers—highly focused laser beams that act as "tractors" to hold and move single particles.
BigTech keeps eating into some of the top Quantum talent whether it’s at Microsoft, Google or elsewhere. As Quantum startups are relegated to a weird purgatory where real Quantum computers don’t exactly exist yet in a commercially viable manner in all seriousness. Obviously there’s a lot of potential here, but so far only IonQ has been able to generate decent revenue as a public company, where Quantinuum will immediately be the leader once it finally goes public. IONQ 0.00%↑ We are still early days for Quantum computing and its long history.
Google Quantum does its fair share of PR, but it’s hard sometimes to find the substance in the research. However the dual approach makes a lot of sense and enables them to collect more talent and build a more tangible approach.
“The mission at our Google Quantum AI lab is to build quantum computing for otherwise unsolvable problems” - Yossi Matias
As vague as these statements can be, Google Quantum appears to be fairly serious in their objectives.
While promising, the Neutral Atom approach technology still faces hurdles:
Laser Precision: Controlling hundreds of lasers with sub-micron accuracy is a massive engineering feat.
Atom Loss: Occasionally, an atom can be “knocked” out of its trap, requiring the system to reload and re-initialize the grid.
Gate Speed: Neutral atom gates are generally slower than superconducting qubits, though they compensate with longer coherence times.
Adam Kaufman you must have realized is fairly young to be given such a role and must be an extraordinary talent. adam.kaufman@colorado.edu
In this YouTube video of around 2024 he describes what his Lab does.
Atomic clocks related to Neutral Atoms also have military and national defense implications. Which ironically is also part of how Inflection claims it generates some revenue. The utility relates to post GPS-navigation and has implications for Space warfare that is certainly coming. These are also important as you can imagine in Submarines and undersea drones and pinpoint navigation.
“This move further embeds our research within one of the most sophisticated physics ecosystems in the world, alongside partners like NIST and QuEra.” - Google Quantum
QuEra is a key partner of Google, thus Google now has as much insight into Neutral Atom computing as possible. Atom Computing was a fairly good team. For more than a decade, Google has focused on superconducting qubits — which are essentially tiny circuits cooled to near absolute zero.
Google is calling this an “experimental hardware team” that will be based in Boulder, Colorado, a region that hosts a significant concentration of AMO research at institutions including CU Boulder, JILA and NIST Boulder.
In general, his research focuses on using alkaline-earth atoms (like Strontium and Ytterbium) as qubits. These systems offer "any-to-any" connectivity, allowing qubits to interact with any other qubit in a large array, which is a significant advantage for error correction compared to fixed-grid superconducting chips.
While no large-scale academic consensus exists as to which qubit approach is the best. Google’s two are certainly I believe are front runners.
A Hybrid Approach - Multi-Modal Qubit Approaches
Google Quantum in some sense is really mirroring the hybrid consensus approach. The overwhelming consensus among leading experts is that the field is multi-modal: different qubit technologies have complementary strengths, and the future of quantum computing will likely involve parallel development, hybridization, and integration rather than a single winner.
Google’s neutral atom research is built on three pillars, which include adaptation of quantum error correction protocols to the physical connectivity of neutral atom arrays and use of high-performance computational modelling and simulation to optimise hardware architecture and error budgets. The third pillar focuses on experimental development of atomic qubit systems at application-relevant scales.
This video features Dr. Adam Kaufman discussing the technical foundations of his research, specifically the use of neutral atoms for both high-precision atomic clocks and quantum computing qubits.
Good luck Adam!