Quantum AI for Pneumonia Detection

Quantum computing in action
Photo Credit: © LMU

Scientific Frontline: Extended "At a Glance" Summary: Quantum AI for Pneumonia Detection

The Core Concept: An AI-assisted medical image analysis model that leverages quantum computing to rapidly and accurately diagnose diseases like pneumonia from X-ray scans.

Key Distinction/Mechanism: Unlike traditional convolutional neural networks (CNNs) that require massive datasets to prevent overfitting, this quantum model learns probability distributions using quantum annealing. It achieves high accuracy (84 to 86 percent) using fewer than 9,000 trainable parameters, compared to the 11 million parameters required by comparable classical systems like ResNet-18.

Major Frameworks/Components:

• Quantum Boltzmann Machines (QBMs): Probabilistic models designed to learn probability distributions directly from training data.
• Quantum Annealing: An optimization technique that exploits quantum mechanical effects, such as quantum tunneling, to drive the sampling process required for training and inference.
• QuCUN Platform: The Quantum Computing User Network, a collaborative platform involving LMU, Aqarios, BASF, and SAP, which hosts the quantum algorithm for real-world testing.

AI in Academic Writing: Enhancing Student Skills

Dr. Emily Dux Speltz, assistant professor in the Department of Humanities and Communication at Embry‑Riddle Worldwide, taught an experimental course that observed and guided students’ experience with AI-assisted writing.
Photo Credit: Christopher Gannon/Iowa State University News Service

Scientific Frontline: Extended "At a Glance" Summary: Generative AI in Academic Writing

The Core Concept: Generative artificial intelligence can serve as a collaborative tool to enhance students' understanding of the writing process, rather than acting as a fully automated replacement for original thought.

Key Distinction/Mechanism: Unlike traditional search queries, writing with AI requires iterative human intervention. Users must carefully design initial prompts, critically evaluate the output for stylistic inconsistencies and factual errors, and revise the text to achieve specific rhetorical objectives.

Major Frameworks/Components: 

• The methodology relies on three "threshold concepts" regarding AI utilization:
• Writing with AI is an experimental process requiring continuous refinement.
• Writing with AI requires human expertise and dialogue to evaluate and guide the output accurately.
• Writing with AI should augment, rather than replace, a student's rhetorical agency.

MIT FloatForm: Self-Assembling Robot Boats

Caption:These small square robotic boats can assemble themselves into larger structures on the water, break apart, and reassemble into something new, all with minimal human direction.
Image Credit: Alex Shipps/MIT CSAIL, using assets from the researchers.

Scientific Frontline: Extended "At a Glance" Summary: FloatForm

The Core Concept: FloatForm is a decentralized swarm of small, self-contained robotic boats that can autonomously assemble, reconfigure, and navigate as a unified floating structure on water.

Key Distinction/Mechanism: Unlike traditional self-assembling systems that rely heavily on a central computer, FloatForm uses a distributed, bio-inspired approach similar to fire ant rafts. A lightweight central planner is used sparingly for final geometric precision, but the robots primarily coordinate locally, allowing the entire swarm to scale and move simultaneously without computational bottlenecks.

Major Frameworks/Components: 

• Decentralized Coordination Algorithm: A localized computing framework where robots coordinate by exchanging positions with immediate neighbors, eliminating the single points of failure found in centralized planning.
• Origami-Inspired Auxetic Latching: An internal, energy-efficient magnetic coupling system driven by a single servo motor. It only consumes power during the act of latching or de-latching, holding its configuration passively via a 3D-printed gearbox.
• Omnidirectional Propulsion: A configuration of four miniature thrusters arranged in an “X” pattern, stabilized by hydrodynamic fins, granting each small vessel precise, multidirectional maneuverability.

Branch of Science: Robotics, Computer Science, Marine Engineering, and Artificial Intelligence.

Future Application: The autonomous assembly of temporary bridges for emergency response, floating infrastructure (such as markets or festival stages), adaptive sensor networks for environmental monitoring, and reconfigurable docking stations in hard-to-reach offshore areas.

Why It Matters: As urban centers become denser, FloatForm transforms static waterways into dynamic, programmable extensions of the city. It offers a highly scalable, resilient method for offloading land-based stress onto underutilized water surfaces.

IRL: LLMs Clarify Vague Robot Commands

“Masked IRL” helps a robot understand ambiguous instructions so it does chores safely. An LLM first elaborates on users' prompts based on demonstration data, then another narrows down which details an algorithm should incorporate into a motion plan.
Image Credit: Gabriel Maragaño

Scientific Frontline: Extended "At a Glance" Summary: Masked Inverse Reinforcement Learning (Masked IRL)

The Core Concept: A machine learning approach that utilizes dual large language models (LLMs) to clarify ambiguous human instructions and filter out irrelevant environmental data, enabling robots to safely execute complex tasks.

Key Distinction/Mechanism: Traditional robotic training requires extensive manual coding or exhaustive physical demonstrations. Masked IRL streamlines this by using one LLM to expand upon vague user prompts based on physical demonstration data, while a second LLM "masks" (ignores) irrelevant environmental details—scoring them as "0"—and prioritizing critical elements as "1" for the final algorithmic motion plan.

Origin/History: Developed by researchers at the Massachusetts Institute of Technology's Computer Science and Artificial Intelligence Laboratory (CSAIL) and slated for presentation at the June 2026 IEEE International Conference on Robotics and Automation.

Optoelectronic Neuromorphic AI Device

Illustration depicts a new phototransistor that integrates light sensing, memory and signal processing.
Image Credit: Courtesy of Oregon State University

Scientific Frontline: Extended "At a Glance" Summary: Programmable Optoelectronic Neuromorphic Device

The Core Concept: Researchers have developed a novel light-sensitive phototransistor that integrates sensing, memory, and signal processing into a single unit. Inspired by the human brain, the device uniquely controls how digital memories strengthen or fade over time.

Key Distinction/Mechanism: Unlike conventional AI hardware that separates sensing and memory components, this device processes information directly at the sensor level. It uses trapped electrical charges from absorbed light as memory and applies an electrical gate voltage to move these charges relative to the transistor channel, actively tuning memory lifetime and decay.

Major Frameworks/Components: 

• Oxide Semiconductor: Functions as the transistor channel to carry electrical current.
• Organic Photosensitive Material: Absorbs light, generates electrical charges, and traps them to form a memory of past optical signals.
• Tunable Charge Positioning: An applied electrical signal adjusts the physical proximity of trapped charges to the microscopic pathway, dictating the persistence or rapid decay of the memory.

Computational Chemistry: In-Depth Description

Computational chemistry is a vital sub-discipline of chemical science that leverages advanced mathematical algorithms, computer software, and theoretical physics to simulate, predict, and analyze molecular structures, dynamic behaviors, and material properties. Its primary goal is to translate the fundamental laws of quantum and classical mechanics into functional computational models. By doing so, it allows scientists to explore complex chemical phenomena that may be too rapid, hazardous, or challenging to observe directly in a laboratory setting, while also guiding experimentalists toward promising discoveries prior to physical synthesis.

PAINT Database: Open Data for Solar Tower Plants

Solar towers in test operation. In Jülich, the DLR operates a large-scale research facility for solar irradiation testing that is unique in Europe.
Photo Credit: German Aerospace Center (DLR)

Scientific Frontline: Extended "At a Glance" Summary: The PAINT Database for Solar Power Tower Plants

The Core Concept: The PAINT database is a freely accessible, FAIR-compliant dataset containing comprehensive operational data from the Jülich Solar Tower test power plant. It provides researchers with real-world information to accelerate the development of more efficient and reliable solar thermal energy generation.

Key Distinction/Mechanism: While photovoltaic systems generate electricity directly, solar towers use movable mirrors (heliostats) to direct sunlight onto a central receiver to generate heat. Operating these systems is highly complex; PAINT bridges the research gap by offering open-source access to 849 gigabytes of structured operational data, allowing engineers to simulate and optimize control mechanisms through digital twins and AI without needing direct access to physical power plants.

Major Frameworks/Components: 

• FAIR Principles: Guiding data formatting to ensure it is Findable, Accessible, Interoperable, and Reusable.
• Spatio-Temporal Asset Catalog (STAC): A standard used to structure spatial and temporal data for optimal human and machine readability.
• Python Integration: Dedicated software that allows researchers to download specific heliostat data and feed it directly into machine-learning models.
• Extensive Metric Repositories: Includes the precise positions, dimensions, and dynamic movements of 2,014 mirrors, alongside weather data, measurements of mirror surface warping, and over 218,000 alignment-verification images.

AI Without Hallucinations: Multi-Agent Protocol

Image Credit: Courtesy of Binghamton University

Scientific Frontline: Extended "At a Glance" Summary: Multi-Agent AI Verification Protocol

The Core Concept: A novel artificial intelligence protocol designed to eliminate hallucinations by forcing multiple large language models (LLMs) to reference authoritative databases and "vote" on the most accurate response.

Key Distinction/Mechanism: Unlike relying on a single generative AI model that might confidently produce false information, this method leverages retrieval-augmented generation (RAG) across multiple open-source chatbots. The models submit their answers for a consensus vote, ensuring the final output is rigorously validated by a majority of the AI agents.

Major Frameworks/Components:

• Retrieval-Augmented Generation (RAG): Forces AI models to consult authoritative medical terminology databases before generating responses.
• Multi-Agent Voting Mechanism: Utilizes an array of open-source LLMs (typically seven per experiment) to cross-verify answers and establish an evidence-based consensus.
• Digital Twins: Dynamic, virtual replicas of physical processes continuously updated with real-time data to create predictive simulations for precision medicine.
• Multi-Scale Network Models: Extracts and verifies evidence across varying data scales, ranging from multiomics to epidemiological and behavioral sources.

Computational Neuroscience: In-Depth Description

Computational neuroscience is the rigorous, interdisciplinary study of brain function in terms of the information processing properties of the nervous system. The primary goal of this field is to understand how electrical and chemical signals are generated, transmitted, and integrated across neurons to produce cognition, perception, and behavior. By constructing theoretical frameworks and employing mathematical models, computational neuroscientists seek to decode the fundamental algorithms of the brain, linking biophysical mechanisms at the cellular level to complex network dynamics.

Stopping AI Model Collapse and Data Cannibalism

Image Credit: Deborah Lupton
(CC BY 4.0)

Scientific Frontline: Extended "At a Glance" Summary: Overcoming AI Data Cannibalism

The Core Concept: AI "Data Cannibalism," also known as Model Collapse, is a phenomenon where artificial intelligence models degrade and produce inaccurate gibberish when continuously trained on synthetic, AI-generated data instead of fresh human data.

Key Distinction/Mechanism: Researchers discovered that integrating just a single real-world data point from outside the closed loop—or incorporating prior knowledge during training—can prevent model collapse entirely, even when the model is overwhelmed by an infinite amount of machine-generated data.

Origin/History: The term "Model Collapse" was first coined in 2024. A foundational breakthrough study detailing its statistical prevention was published in Physical Review Letters in May 2026 by researchers from King's College London, the Norwegian University of Science and Technology, and the Abdus Salam International Centre for Theoretical Physics.

Improving the reliability of circuits for quantum computers

This illustration uses a layered sculpture to interpret a phenomenon that can cause a quantum circuit to perform differently than expected, increasing the error in computations. MIT researchers developed a method to detect and precisely measure the strength of these distortions.
Image Credit: Amy Pan and Sampson Wilcox
(CC BY-NC-ND 3.0)

Scientific Frontline: Extended "At a Glance" Summary: Quantum Circuit Reliability via Harmonic Detection

The Core Concept: A novel diagnostic technique enables the detection and precise measurement of "second-order harmonic corrections," a non-linear distortion that causes superconducting quantum circuits to deviate from expected operational behaviors.

Key Distinction/Mechanism: Functional superconducting circuits rely on Cooper pairs of electrons quantum tunneling through a Josephson junction barrier one pair at a time. Second-order harmonic corrections occur when two pairs tunnel simultaneously. This two-pair tunneling, driven by additional inductance from connective wiring rather than the junction's intrinsic dynamics, bypasses the circuit's intended single-pair limitations.

Major Frameworks/Components:

• Josephson Junctions: Critical circuit elements consisting of two superconducting wires separated by a nanometer-scale barrier, enabling the transfer and manipulation of quantum information.
• Cooper Pairs: Paired charge-carrying electrons that transport current via quantum tunneling.
• Second-Order Harmonic Corrections: The specific distortion caused by the simultaneous multi-pair tunneling effect.
• Series Inductance: The tendency of wires to oppose changes in electric current flow, identified as the primary source of these harmonic distortions in the tested devices.

UC Irvine-led study achieves brain-controlled walking with artificial sensory feedback

UC Irvine researchers (from left) Dr. An Do, associate professor of neurology; Payam Heydari, professor of electrical engineering and computer science; and Zoran Nenadic, professor of biomedical engineering, recently participated in a study that demonstrated a brain-computer interface technology that enables spinal cord injury patients to walk with a robotic exoskeleton and feel lifelike sensory responses, a key factor in safe and realistic mobility.
Photo Credit: Debbie Morales / UC Irvine

Scientific Frontline: Extended "At a Glance" Summary: Bidirectional Brain-Computer Interface for Walking

The Core Concept: A bidirectional brain-computer interface (BDBCI) that enables individuals to control a robotic walking exoskeleton using brain signals while simultaneously receiving artificial leg sensation through direct electrical stimulation of the sensory cortex.

Key Distinction/Mechanism: Unlike existing robotic exoskeletons that rely on manual control and lack sensory feedback, this system decodes motor intent from electrocorticography (ECoG) signals in the leg motor cortex and delivers real-time artificial sensation to the somatosensory cortex. This bidirectional approach creates a closed-loop, brain-driven walking experience, which improves gait speed and reduces the risk of falls.

Major Frameworks/Components:

• Bidirectional Brain-Computer Interface (BDBCI): An embedded, portable platform utilizing high-speed microcontrollers for neural signal acquisition, real-time decoding, electrical stimulation, and wireless communication without relying on a tethered computer.
• Bilateral Interhemispheric Electrocorticography (ECoG): Implants strategically placed to access the leg motor and sensory cortices within the medial wall of the brain along the interhemispheric fissure.
• Direct Cortical Electrical Stimulation: A localized technique used to safely and practically elicit artificial sensory feedback directly in the somatosensory cortex.
• Robotic Gait Exoskeleton: Integration with a powered exoskeleton to translate decoded brain signals into physical, bilateral lower-extremity movement.

Smart cable sharing gives quantum computers a big boost

An artist’s rendering of time multiplexing of control signals to a quantum computer. The control signals for single-qubit gates (short blocks) and two-qubit gates (long blocks) travel through common cables (tunnels) to switches, which distribute them among the qubits (spheres) based on switching signals (diamonds). By ordering the control signals in a clever way, akin to playing Tetris, traffic jams in the flow of control signals can largely be avoided and programs on the quantum computer can be executed almost as fast as if each qubit had its own cable for control signals.
Image Credit: Chalmers University of Technology/Boid

Scientific Frontline: Extended "At a Glance" Summary: Smart Cable Sharing in Quantum Computing

The Core Concept: Smart cable sharing (time-domain multiplexing) is a control architecture that allows multiple qubits to be operated sequentially via a single shared cable. This drastically reduces internal hardware requirements without significantly slowing down the system's computation time.

Key Distinction/Mechanism: In traditional quantum computing architectures, each qubit requires its own dedicated control cable (parallel control), which generates excess heat and takes up physical space. Smart cable sharing functions differently by utilizing time-domain multiplexing; it routes rapid, sequential control signals through shared cables down to microwave switches located directly next to the quantum processor to direct the signals to the correct target qubits.

Major Frameworks/Components:

• Superconducting Circuits: The foundational quantum hardware that must be cooled inside cryostats to near absolute zero (-273.15°C) to function properly.
• Time-Domain Multiplexing: The technique of sequencing control signals rapidly so that qubits do not require simultaneous, dedicated input.
• Microwave Switches: Rapid routing mechanisms installed directly next to the processor to distribute shared signals to individual qubits.
• Logarithmic Time Scaling: A critical mathematical finding from the research demonstrating that computational delay increases logarithmically—not linearly—as the number of qubits sharing a cable increases.

Living Brain Cells Enable Machine Learning Computations

(a) Conventional neuron models used in reservoir computing. Artificial neural networks (ANNs) comprise of neuron models that sum up weighted inputs, filter the value through an activation function, and generate a continuous valued output. Spiking neural networks (SNNs) comprise of neuron models receive spiking inputs and output spikes when their membrane potential exceeds a threshold. (b) Biological neurons used for reservoir computing in this work. Rat cortical neurons are cultured in microfluidic devices that are attached to a microelectrode array.
Image Credit: ©Yuki Sono et al.

Scientific Frontline: Extended "At a Glance" Summary: Living Brain Cells Enable Machine Learning Computations

The Core Concept: Biological neural networks (BNNs) grown from cultured neurons can be integrated into a machine learning framework to perform supervised temporal pattern learning. This demonstrates that living cellular systems can generate complex, time-series computations previously restricted to artificial systems.

Key Distinction/Mechanism: Unlike traditional artificial neural networks (ANNs) or spiking neural networks (SNNs) that rely on digital models of neurons, this system utilizes living rat cortical neurons cultured on microelectrode arrays within microfluidic devices. By applying the First-Order Reduced and Controlled Error (FORCE) learning algorithm to this "physical reservoir," researchers optimized the readout layer to correct errors in real-time, enabling the living network to generate structured temporal signals such as sine waves and chaotic trajectories.

Major Frameworks/Components:

• Reservoir Computing: A computational framework that processes time-dependent data by leveraging the dynamic properties of complex, recurrently connected networks.
• FORCE Learning: A real-time adaptation technique used to train the system by continuously adjusting output signals in response to real-time feedback errors.
• Microfluidic Network Architecture: Specialized devices used to guide biological neuronal growth and control connectivity, promoting the high-dimensional dynamics required for computation by minimizing excessive neural synchronization.
• Biological Neural Networks (BNNs): The living substrate of cultured rat cortical neurons that functions as the core processing reservoir.

AI sheds light on an ancient gaming mystery

Above: the possible gameboard with pencil marks highlighting the incised lines. Below: diagram of the lines, indicating how pieces may have been moved along them to play the game
Image Credit: Walter Crist

Scientific Frontline: "At a Glance" Summary: AI Decoding of an Ancient Roman Board Game

• Main Discovery: Researchers successfully utilized artificial intelligence to decode the rules of an ancient, previously unexplainable board game carved into a limestone object discovered in the Roman Netherlands.
• Methodology: The research team employed the AI-driven play system Ludii to simulate hundreds of rule sets from documented ancient European games, systematically adjusting parameters to identify which simulated movements replicated the specific, asymmetrical wear patterns observed on the original artifact.
• Key Data: The AI simulations consistently reproduced the concentrated friction and uneven wear along the carved lines when applying rules for a "blocking game," characterized by asymmetrical play where a player with more pieces attempts to trap an opponent with fewer pieces.
• Significance: This study represents the first successful integration of AI-driven simulated play with archaeological analysis to identify a board game, providing physical evidence that blocking games existed long before their earliest prior documentation in the Middle Ages.
• Future Application: This computational approach establishes a new analytical framework for archaeologists to interpret mysterious historical artifacts and reconstruct undocumented cultural practices when written texts or artworks have not survived.
• Branch of Science: Archaeology, Computer Science, and Cultural History.
• Additional Detail: The artifact provided a rare preservation opportunity, as most everyday Roman games were historically drawn in dust or carved into perishable materials like wood, leaving minimal physical evidence for modern physical analysis.

Engineers design structures that compute with heat

This artistic rendering shows a thermal analog computing device, which performs computations using excess heat, embedded in a microelectronic system.
Image Credit: Jose-Luis Olivares, MIT
(CC BY-NC-ND 4.0)

Scientific Frontline: "At a Glance" Summary

• Main Discovery: Researchers have developed microscopic silicon structures capable of performing analog computations by utilizing waste heat instead of electricity.
• Methodology: The team employed an "inverse design" software system to iteratively optimize the geometry and porosity of silicon metastructures, enabling them to conduct and diffuse heat in specific patterns that represent mathematical operations.
• Key Data: The thermal computing structures achieved over 99 percent accuracy in performing matrix-vector multiplications, a fundamental calculation for machine learning models.
• Significance: This paradigm shifts heat from a problematic waste product to a functional information carrier, potentially allowing for energy-free thermal sensing and signal processing within microelectronics.
• Future Application: Beyond thermal management, the technology is envisioned for use in sequential machine learning operations and programmable thermal structures that can detect localized heat gradients without digital components.
• Branch of Science: Mechanical Engineering, Applied Physics, and Computer Science.
• Additional Detail: To handle negative numerical values—which heat conduction cannot naturally represent—the researchers developed a method to split matrices into positive and negative components, optimizing separate structures for each.

Efficient cooling method could enable chip-based quantum computers

Caption:Researchers developed a photonic chip that incorporates precisely designed antennas to manipulate beams of tightly focused, intersecting light, which can rapidly cool a quantum computing system to someday enable greater efficiency and stability.
Illustration Credit: Michael Hurley and Sampson Wilcox
(CC BY-NC-ND 4.0)

Scientific Frontline: "At a Glance" Summary

• Core Discovery: Researchers successfully demonstrated a high-efficiency polarization-gradient cooling method integrated directly onto a photonic chip, enabling faster and more effective cooling for trapped-ion quantum computers.
• Methodology: The system utilizes precisely designed nanoscale antennas connected by waveguides to emit intersecting light beams with diverse polarizations; this creates a rotating light vortex that drastically reduces the kinetic energy of trapped ions.
• Key Data: The approach achieved ion cooling temperatures nearly 10 times below the standard Doppler limit, reaching this state in approximately 100 microseconds—several times faster than comparable techniques.
• Context: Unlike traditional quantum setups that rely on bulky external lasers and are sensitive to vibrations, this integrated architecture generates stable optical fields directly on the chip, eliminating the need for complex external optical alignment.
• Significance: This advancement validates a scalable path for quantum computing where thousands of ion-interface sites can coexist on a single chip, significantly improving the stability and practicality of quantum information processing.
• Specific Mechanism: The on-chip antennas feature specialized curved notches designed to scatter light upward, maximizing the optical interaction with the ion and allowing for advanced operations beyond simple cooling.

Nature-inspired computers are shockingly good at math

Researchers Brad Theilman, center, and Felix Wang, behind, unpack a neuromorphic computing core at Sandia National Laboratories. While the hardware might look similar to a regular computer, the circuitry is radically different. It applies elements of neuroscience to operate more like a brain, which is extremely energy-efficient.
Photo Credit: Craig Fritz

Scientific Frontline: "At a Glance" Summary

• Main Discovery: Neuromorphic (brain-inspired) computing systems have been proven capable of solving partial differential equations (PDEs) with high efficiency, a task previously believed to be the exclusive domain of traditional, energy-intensive supercomputers.
• Methodology: Researchers at Sandia National Laboratories developed a novel algorithm that utilizes a circuit model based on cortical networks to execute complex mathematical calculations, effectively mapping brain-like architecture to rigorous physical simulations.
• Theoretical Breakthrough: The study establishes a mathematical link between a computational neuroscience model introduced 12 years ago and the solution of PDEs, demonstrating that neuromorphic hardware can handle deterministic math, not just pattern recognition.
• Comparison: Unlike conventional supercomputers that require immense power for simulations (such as fluid dynamics or electromagnetic fields), this neuromorphic approach mimics the brain's ability to perform exascale-level computations with minimal energy consumption.
• Primary Implication: This advancement could enable the development of neuromorphic supercomputers for national security and nuclear stockpile simulations, significantly reducing the energy footprint of critical scientific modeling.
• Secondary Significance: The findings suggest that "diseases of the brain could be diseases of computation," providing a new framework for understanding neurological conditions by studying how these biological-style networks process information.

The Quest for the Synthetic Synapse

Spike Timing" difference (Biology vs. Silicon)
Image Credit: Scientific Frontline

The modern AI revolution is built on a paradox: it is incredibly smart, but thermodynamically reckless. A large language model requires megawatts of power to function, whereas the human brain—which allows you to drive a car, debate philosophy, and regulate a heartbeat simultaneously—runs on roughly 20 watts, the equivalent of a dim lightbulb.

To close this gap, science is moving away from the "Von Neumann" architecture (where memory and processing are separate) toward Neuromorphic Computing—chips that mimic the physical structure of the brain. This report analyzes how close we are to building a "synthetic synapse."

Breakthrough could connect quantum computers at 200 times longer distance

New research from University of Chicago Pritzker School of Molecular Engineering Asst. Prof. Tian Zhong could make it possible for quantum computers to connect at distances up to 1,243 miles, shattering previous records.
Photo Credit: Jason Smith

A new nanofabrication approach could increase the range of quantum networks from a few kilometers to a potential 2,000 km, bringing quantum internet closer than ever

Quantum computers are powerful, lightning-fast and notoriously difficult to connect to one another over long distances. 

Previously, the maximum distance two quantum computers could connect through a fiber cable was a few kilometers. This means that, even if such cable were run between them, quantum computers in downtown Chicago’s Willis Tower and the University of Chicago Pritzker School of Molecular Engineering (UChicago PME) on the South Side would be too far apart to communicate with each other.