Tech companies have been funneling billions of dollars into quantum computers for years. The hope is that they’ll be a game changer for fields as diverse as finance, drug discovery, and logistics.

Those expectations have been especially high in physics and chemistry, where the weird effects of quantum mechanics come into play. In theory, this is where quantum computers could have a huge advantage over conventional machines.

But while the field struggles with the realities of tricky quantum hardware, another challenger is making headway in some of these most promising use cases. AI is now being applied to fundamental physics, chemistry, and materials science in a way that suggests quantum computing’s purported home turf might not be so safe after all.

The scale and complexity of quantum systems that can be simulated using AI is advancing rapidly, says Giuseppe Carleo, a professor of computational physics at the Swiss Federal Institute of Technology (EPFL). Last month, he coauthored a paper published in Science showing that neural-network-based approaches are rapidly becoming the leading technique for modeling materials with strong quantum properties. Meta also recently unveiled an AI model trained on a massive new data set of materials that has jumped to the top of a leaderboard for machine-learning approaches to material discovery.

Given the pace of recent advances, a growing number of researchers are now asking whether AI could solve a substantial chunk of the most interesting problems in chemistry and materials science before large-scale quantum computers become a reality. 

“The existence of these new contenders in machine learning is a serious hit to the potential applications of quantum computers,” says Carleo “In my opinion, these companies will find out sooner or later that their investments are not justified.”

Exponential problems

The promise of quantum computers lies in their potential to carry out certain calculations much faster than conventional computers. Realizing this promise will require much larger quantum processors than we have today. The biggest devices have just crossed the thousand-qubit mark, but achieving an undeniable advantage over classical computers will likely require tens of thousands, if not millions. Once that hardware is available, though, a handful of quantum algorithms, like the encryption-cracking Shor’s algorithm, have the potential to solve problems exponentially faster than classical algorithms can. 

But for many quantum algorithms with more obvious commercial applications, like searching databases, solving optimization problems, or powering AI, the speed advantage is more modest. And last year, a paper coauthored by Microsoft’s head of quantum computing, Matthias Troyer, showed that these theoretical advantages disappear if you account for the fact that quantum hardware operates orders of magnitude slower than modern computer chips. The difficulty of getting large amounts of classical data in and out of a quantum computer is also a major barrier. 

So Troyer and his colleagues concluded that quantum computers should instead focus on problems in chemistry and materials science that require simulation of systems where quantum effects dominate. A computer that operates along the same quantum principles as these systems should, in theory, have a natural advantage here. In fact, this has been a driving idea behind quantum computing ever since the renowned physicist Richard Feynman first proposed the idea.

The rules of quantum mechanics govern many things with huge practical and commercial value, like proteins, drugs, and materials. Their properties are determined by the interactions of their constituent particles, in particular their electrons—and simulating these interactions in a computer should make it possible to predict what kinds of characteristics a molecule will exhibit. This could prove invaluable for discovering things like new medicines or more efficient battery chemistries, for example. 

But the intuition-defying rules of quantum mechanics—in particular, the phenomenon of entanglement, which allows the quantum states of distant particles to become intrinsically linked—can make these interactions incredibly complex. Precisely tracking them requires complicated math that gets exponentially tougher the more particles are involved. That can make simulating large quantum systems intractable on classical machines.

This is where quantum computers could shine. Because they also operate on quantum principles, they are able to represent quantum states much more efficiently than is possible on classical machines. They could also take advantage of quantum effects to speed up their calculations.

But not all quantum systems are the same. Their complexity is determined by the extent to which their particles interact, or correlate, with each other. In systems where these interactions are strong, tracking all these relationships can quickly explode the number of calculations required to model the system. But in most that are of practical interest to chemists and materials scientists, correlation is weak, says Carleo. That means their particles don’t affect each other’s behavior significantly, which makes the systems far simpler to model.

The upshot, says Carleo, is that quantum computers are unlikely to provide any advantage for most problems in chemistry and materials science. Classical tools that can accurately model weakly correlated systems already exist, the most prominent being density functional theory (DFT). The insight behind DFT is that all you need to understand a system’s key properties is its electron density, a measure of how its electrons are distributed in space. This makes for much simpler computation but can still provide accurate results for weakly correlated systems.

Simulating large systems using these approaches requires considerable computing power. But in recent years there’s been an explosion of research using DFT to generate data on chemicals, biomolecules, and materials—data that can be used to train neural networks. These AI models learn patterns in the data that allow them to predict what properties a particular chemical structure is likely to have, but they are orders of magnitude cheaper to run than conventional DFT calculations. 

This has dramatically expanded the size of systems that can be modeled—to as many as 100,000 atoms at a time—and how long simulations can run, says Alexandre Tkatchenko, a physics professor at the University of Luxembourg. “It’s wonderful. You can really do most of chemistry,” he says.

Olexandr Isayev, a chemistry professor at Carnegie Mellon University, says these techniques are already being widely applied by companies in chemistry and life sciences. And for researchers, previously out of reach problems such as optimizing chemical reactions, developing new battery materials, and understanding protein binding are finally becoming tractable.

As with most AI applications, the biggest bottleneck is data, says Isayev. Meta’s recently released materials data set was made up of DFT calculations on 118 million molecules. A model trained on this data achieved state-of-the-art performance, but creating the training material took vast computing resources, well beyond what’s accessible to most research teams. That means fulfilling the full promise of this approach will require massive investment.

Modeling a weakly correlated system using DFT is not an exponentially scaling problem, though. This suggests that with more data and computing resources, AI-based classical approaches could simulate even the largest of these systems, says Tkatchenko. Given that quantum computers powerful enough to compete are likely still decades away, he adds, AI’s current trajectory suggests it could reach important milestones, such as precisely simulating how drugs bind to a protein, much sooner.

Strong correlations

When it comes to simulating strongly correlated quantum systems—ones whose particles interact a lot—methods like DFT quickly run out of steam. While more exotic, these systems include materials with potentially transformative capabilities, like high-temperature superconductivity or ultra-precise sensing. But even here, AI is making significant strides.

In 2017, EPFL’s Carleo and Microsoft’s Troyer published a seminal paper in Science showing that neural networks could model strongly correlated quantum systems. The approach doesn’t learn from data in the classical sense. Instead, Carleo says, it is similar to DeepMind’s AlphaZero model, which mastered the games of Go, chess, and shogi using nothing more than the rules of each game and the ability to play itself.

In this case, the rules of the game are provided by Schrödinger’s equation, which can precisely describe a system’s quantum state, or wave function. The model plays against itself by arranging particles in a certain configuration and then measuring the system’s energy level. The goal is to reach the lowest energy configuration (known as the ground state), which determines the system’s properties. The model repeats this process until energy levels stop falling, indicating that the ground state—or something close to it—has been reached.

The power of these models is their ability to compress information, says Carleo. “The wave function is a very complicated mathematical object,” he says. “What has been shown by several papers now is that [the neural network] is able to capture the complexity of this object in a way that can be handled by a classical machine.”

Since the 2017 paper, the approach has been extended to a wide range of strongly correlated systems, says Carleo, and results have been impressive. The Science paper he published with colleagues last month put leading classical simulation techniques to the test on a variety of tricky quantum simulation problems, with the goal of creating a benchmark to judge advances in both classical and quantum approaches.

Carleo says that neural-network-based techniques are now the best approach for simulating many of the most complex quantum systems they tested. “Machine learning is really taking the lead in many of these problems,” he says.

These techniques are catching the eye of some big players in the tech industry. In August, researchers at DeepMind showed in a paper in Science that they could accurately model excited states in quantum systems, which could one day help predict the behavior of things like solar cells, sensors, and lasers. Scientists at Microsoft Research have also developed an open-source software suite to help more researchers use neural networks for simulation.

One of the main advantages of the approach is that it piggybacks on massive investments in AI software and hardware, says Filippo Vicentini, a professor of AI and condensed-matter physics at École Polytechnique in France, who was also a coauthor on the Science benchmarking paper: “Being able to leverage these kinds of technological advancements gives us a huge edge.”

There is a caveat: Because the ground states are effectively found through trial and error rather than explicit calculations, they are only approximations. But this is also why the approach could make progress on what has looked like an intractable problem, says Juan Carrasquilla, a researcher at ETH Zurich, and another coauthor on the Science benchmarking paper.

If you want to precisely track all the interactions in a strongly correlated system, the number of calculations you need to do rises exponentially with the system’s size. But if you’re happy with an answer that is just good enough, there’s plenty of scope for taking shortcuts. 

“Perhaps there’s no hope to capture it exactly,” says Carrasquilla. “But there’s hope to capture enough information that we capture all the aspects that physicists care about. And if we do that, it’s basically indistinguishable from a true solution.”

And while strongly correlated systems are generally too hard to simulate classically, there are notable instances where this isn’t the case. That includes some systems that are relevant for modeling high-temperature superconductors, according to a 2023 paper in Nature Communications.

“Because of the exponential complexity, you can always find problems for which you can’t find a shortcut,” says Frank Noe, research manager at Microsoft Research, who has led much of the company’s work in this area. “But I think the number of systems for which you can’t find a good shortcut will just become much smaller.”

No magic bullets

However, Stefanie Czischek, an assistant professor of physics at the University of Ottawa, says it can be hard to predict what problems neural networks can feasibly solve. For some complex systems they do incredibly well, but then on other seemingly simple ones, computational costs balloon unexpectedly. “We don’t really know their limitations,” she says. “No one really knows yet what are the conditions that make it hard to represent systems using these neural networks.”

Meanwhile, there have also been significant advances in other classical quantum simulation techniques, says Antoine Georges, director of the Center for Computational Quantum Physics at the Flatiron Institute in New York, who also contributed to the recent Science benchmarking paper. “They are all successful in their own right, and they are also very complementary,” he says. “So I don’t think these machine-learning methods are just going to completely put all the other methods out of business.”

Quantum computers will also have their niche, says Martin Roetteler, senior director of quantum solutions at IonQ, which is developing quantum computers built from trapped ions. While he agrees that classical approaches will likely be sufficient for simulating weakly correlated systems, he’s confident that some large, strongly correlated systems will be beyond their reach. “The exponential is going to bite you,” he says. “There are cases with strongly correlated systems that we cannot treat classically. I’m strongly convinced that that’s the case.”

In contrast, he says, a future fault-tolerant quantum computer with many more qubits than today’s devices will be able to simulate such systems. This could help find new catalysts or improve understanding of metabolic processes in the body—an area of interest to the pharmaceutical industry.

Neural networks are likely to increase the scope of problems that can be solved, says Jay Gambetta, who leads IBM’s quantum computing efforts, but he’s unconvinced they’ll solve the hardest challenges businesses are interested in.

“That’s why many different companies that essentially have chemistry as their requirement are still investigating quantum—because they know exactly where these approximation methods break down,” he says.

Gambetta also rejects the idea that the technologies are rivals. He says the future of computing is likely to involve a hybrid of the two approaches, with quantum and classical subroutines working together to solve problems. “I don’t think they’re in competition. I think they actually add to each other,” he says.

But Scott Aaronson, who directs the Quantum Information Center at the University of Texas, says machine-learning approaches are directly competing against quantum computers in areas like quantum chemistry and condensed-matter physics. He predicts that a combination of machine learning and quantum simulations will outperform purely classical approaches in many cases, but that won’t become clear until larger, more reliable quantum computers are available.

“From the very beginning, I’ve treated quantum computing as first and foremost a scientific quest, with any industrial applications as icing on the cake,” he says. “So if quantum simulation turns out to beat classical machine learning only rarely, I won’t be quite as crestfallen as some of my colleagues.”

One area where quantum computers look likely to have a clear advantage is in simulating how complex quantum systems evolve over time, says EPFL’s Carleo. This could provide invaluable insights for scientists in fields like statistical mechanics and high-energy physics, but it seems unlikely to lead to practical uses in the near term. “These are more niche applications that, in my opinion, do not justify the massive investments and the massive hype,” Carleo adds.

Nonetheless, the experts MIT Technology Review spoke to said a lack of commercial applications is not a reason to stop pursuing quantum computing, which could lead to fundamental scientific breakthroughs in the long run.

“Science is like a set of nested boxes—you solve one problem and you find five other problems,” says Vicentini. “The complexity of the things we study will increase over time, so we will always need more powerful tools.”

There’s little doubt that some of the most important pillars of modern cryptography will tumble spectacularly once quantum computing, now in its infancy, matures sufficiently. Some experts say that could be in the next couple decades. Others say it could take longer. No one knows.

The uncertainty leaves a giant vacuum that can be filled with alarmist pronouncements that the world is close to seeing the downfall of cryptography as we know it. The false pronouncements can take on a life of their own as they’re repeated by marketers looking to peddle post-quantum cryptography snake oil and journalists tricked into thinking the findings are real. And a new episode of exaggerated research has been playing out for the past few weeks.

All aboard the PQC hype train

The last time the PQC—short for post-quantum cryptography—hype train gained this much traction was in early 2023, when scientists presented findings that claimed, at long last, to put the quantum-enabled cracking of the widely used RSA encryption scheme within reach. The claims were repeated over and over, just as claims about research released in September have for the past three weeks.

Read full article

Comments

© Getty Images

It turns out that you don’t need to be a scientist to encode data in DNA. Researchers have been working on DNA-based data storage for decades, but a new template-based method inspired by our cells’ chemical processes is easy enough for even nonscientists to practice. The technique could pave the way for an unusual but ultra-stable way to store information. 

The idea of storing data in DNA was first proposed in the 1950s by the physicist Richard Feynman. Genetic material has exceptional storage density and durability; a single gram of DNA can store a trillion gigabytes of data and retain the information for thousands of years. Decades later, a team led by George Church at Harvard University put the idea into practice, encoding a 53,400-word book.

This early approach relied on DNA synthesis—stringing genetic sequences together piece by piece, like beads on a thread, using the four nucleotide building blocks A, T, C, and G to encode information. The process was expensive, time consuming, and error prone, creating only one bit (or an eighth of a byte) with each nucleotide added to a strand. Crucially, the process required skilled expertise to carry out.

The new method, published in Nature last week, is more efficient, storing 350 bits at a time by encoding strands in parallel. Rather than hand-threading each DNA strand, the team assembles strands from pre-built DNA bricks about 20 nucleotides long, encoding information by altering some and not others along the way. Peking University’s Long Qian and team got the idea for such templates from the way cells share the same basic set of genes but behave differently in response to chemical changes in DNA strands. “Every cell in our bodies has the same genome sequence, but genetic programming comes from modifications to DNA. If life can do this, we can do this,” she says. 

Qian and her colleagues encoded data through methylation, a chemical reaction that switches genes on and off by attaching a methyl compound—a small methane-related molecule. Once the bricks are locked into their assigned spots on the strand, researchers select which bricks to methylate, with the presence or absence of the modification standing in for binary values of 0 or 1. The information can then be deciphered using nanopore sequencers to detect whether a brick has been methylated. In theory, the new method is simple enough to be carried out without detailed knowledge of how to manipulate DNA.

The storage capacity of each DNA strand caps off at roughly 70 bits. For larger files, researchers splintered data into multiple strands identified by unique barcodes encoded in the bricks. The strands were then read simultaneously and sequenced according to their barcodes. With this technique, researchers encoded the image of a tiger rubbing from the Han dynasty, troubleshooting the encoding process until the image came back with no errors. The same process worked for more complex images, like a photorealistic print of a panda. 

To gauge the real-world applicability of their approach, the team enlisted 60 students from diverse academic backgrounds—not just scientists—to encode any writing of their choice. The volunteers transcribed their writing into binary code through a web server. Then, with a kit sent by the team, they pipetted an enzyme into a 96-well plate of the DNA bricks, marking which would be methylated. The team then ran the samples through a sequencer to make the DNA strand. Once the computer received the sequence, researchers ran a decoding algorithm and sent the restored message back to a web server for students to retrieve with a password. The writing came back with a 1.4% error rate in letters, and the errors were eventually corrected through language-learning models. 

Once it’s more thoroughly developed, Qian sees the technology becoming useful as long-term storage for archival information that isn’t accessed every day, like medical records, financial reports, or scientific data.  

The success nonscientists achieved using the technique in coding trials suggests that the DNA storage could eventually become a practical technology. “Everyone is storing data every day, and so to compete with traditional data storage technologies, DNA methods need to be usable by the everyday person,” says Jeff Nivala, co-director of University of Washington’s Molecular Information Systems Lab. “This is still an early demonstration of going toward nonexperts, but I think it’s pretty unique that they’re able to do that.”

DNA storage still has many strides left to make before it can compete with traditional data storage. The new system is more expensive than either traditional data storage techniques or previous DNA-synthesis methods, Nivala says, though the encoding process could become more efficient with automation on a larger scale. With future development, template-based DNA storage might become a more secure method of tackling ever-climbing data demands. 

On Monday, Microsoft came out guns blazing, posting a blog accusing Google of "dishonestly" funding groups conducting allegedly biased studies to discredit Microsoft and mislead antitrust enforcers and the public.

In the blog, Microsoft lawyer Rima Alaily alleged that an astroturf group called the Open Cloud Coalition will launch this week and will appear to be led by "a handful of European cloud providers." In actuality, however, those smaller companies were secretly recruited by Google, which allegedly pays them "to serve as the public face" and "obfuscate" Google's involvement, Microsoft's blog said. In return, Google likely offered the cloud providers cash or discounts to join, Alaily alleged.

The Open Cloud Coalition is just one part of a "pattern of shadowy campaigns" that Google has funded, both "directly and indirectly," to muddy the antitrust waters, Alaily alleged. The only other named example that Alaily gives while documenting this supposed pattern is the US-based Coalition for Fair Software Licensing (CFSL), which Alaily said has attacked Microsoft's cloud computing business in the US, the United Kingdom, and the European Union.

Read full article

Comments

© lcva2 | iStock Editorial / Getty Images Plus

A tech industry leader shares perspective on charting a course forward.

GUEST COLUMN | by Erik Stromquist

As a Chromebook OEM, CTL sits at the center of the edtech ecosystem. In a single day, I may chat with a district CIO at a business development event, take a support call from a tech director, and interface with the Google ChromeOS team. This gives me a unique perspective on what the industry is talking about—from new technology innovation to IT admin requests.

Lately, I’ve seen a few common threads running through many of our recent conversations: cybersecurity, connectivity, and sustainability.  Here is where I see the industry trajectories on all three topics, and how we’re encouraging the industry to work together to solve these common challenges.

Cybersecurity: An Ever-Evolving Challenge

School cybersecurity is increasingly under threat. From ransomware attacks to phishing scams, malicious actors increasingly target schools. In fact, according to K12 SIX data, there have been more than 1,600 attacks on schools since 2016, and unfortunately, no one expects it to get any better soon.

Even though we’re not directly involved in this layer of technology, as a Chromebook manufacturer and edtech solution provider, we are increasingly concerned for our customers. We’ve dedicated our company to ChromeOS cloud computing for one simple reason: ChromeOS has never been hacked. It’s the most secure, containerized operating system that provides the most out-of-the-box protection to users.

However, that’s just the protection out of the box. It’s a solid start, but maintaining that high level of security is critical once a device is deployed and used daily. That’s why, although we’re a hardware manufacturer, we’re investigating new partnerships with fellow travelers in this space. We want to help solve our customer’s cybersecurity concerns together. IT admins need to know the best tools and best practices available to help them prevent breaches and protect student and employee data.

It’s Cybersecurity Month, so this concern is top-of-mind for CTL and our customers. In a few weeks we’re bringing together our IT experts, the ChromeOS Team, a grant funding expert, and a phishing training software provider in a webinar to provide a wealth of cybersecurity updates and vital information to the edtech community.

Connectivity

Bridging the digital divide and closing the homework gap are great initiatives that all depend on the availability of connectivity. CTL believes strongly in providing digital access for every student, regardless of home internet status. We were the first to launch an LTE-connected Chromebook in 2018, and in the next several months we’ll be the first to launch 5G connectivity on Chromebooks for LTE-enablement and private wireless network access. This cause is one of our core corporate pillars of innovation.

The recent changes in E-Rate have left a lot of schools wondering how they can provide digital access beyond the school walls to close that homework gap. The FCC recently ruled it would continue to fund connectivity; however, it is exclusively limited to hotspots. Hotspots are certainly one way to deliver connectivity, but what we hear from our customers and large districts like the Los Angeles Unified School District is that they would prefer the FCC remove the hotspot requirement. Hotspots can be problematic for IT directors to roll out and manage – from provisioning the devices to keeping track of them, from frequent battery replacement to preventing unauthorized users. The hotspot is a limiting device, and we’ve written a letter supporting the LAUSD petition to remove the hotspot-only requirement for funding. School IT departments can select the best-connected devices for their populations if the funding transitions to be device-agnostic. For many school districts, deploying LTE-enabled Chromebooks is a single, streamlined solution that significantly reduces the extra device cost and management time.

Looking forward, CTL is most excited about district-wide private wireless networks for schools and the new 5G-enabled Chromebooks we’ll introduce next year. We’re involved in many districts around the country that are seeking to enable digital access for several use cases, including kids in need, rural communities, and even home-insecure students. Providing hardware and digital access for all students is the last mile in finally conquering the digital learning divide and providing true educational equity. We’re excited by the positive impacts on teaching and learning.

Sustainability

With the proliferation of student laptops, the industry has become increasingly concerned about sustainability over the last several years. Questions often arise, and when they do, we work to broaden the conversation beyond the simple recyclability of components. Sure, recycling is important, however, there is so much more we can do. This mission is so core to CTL that we recently became the first Chromebook manufacturer certified as a B Corporation™.

Going forward, CTL postulates that complete circularity in the Chromebook space is not only possible but a mandate. On the manufacturing side, we examine everything from the amount of post-consumer recycled plastic in our products to the to the Forest Stewardship Council (FSc) certification of our boxes as forest-based materials that meet the highest production standards from a sustainably managed forest. We encourage trade-ins once a district is ready to replace some of its fleet. This enables us to provide a rebate on the old devices towards new purchases, but more than that, it enables us to refurbish devices for a second life rather than simply recycling. This keeps components out of landfills longer and provides additional digital teaching and learning access for new use cases, such as loaners, spares, summer school students, and substitute teachers – the list is expansive.

The industry needs partners to help manage the entire lifecycle of Chromebooks and other electronic devices. Working together, we’re excited to see progress in creating innovative sustainability solutions that are great for learning and the planet.  

Moving Into the Future

CTL is growing by investing in these three key trends for student and teacher computing: cybersecurity, connectivity, and sustainability. I myself am moving on to a new role here at CTL, where I’ll be putting together the strategies, programs, and partnerships to ensure we solve these challenges and provide greater teaching and learning opportunities for educators and students throughout the next decade. If you have ideas or questions, please feel free to reach out to me directly. I’ll be listening.

—

Erik Stromquist is Co-Founder and Chairman of the Board at CTL, a mission-driven company empowering success at school and work with innovative cloud-computing products and industry-leading services. Connect with Erik via LinkedIn. 

The post The Future of School Computing: Three Top Trends appeared first on EdTech Digest.

Lab scientists spend much of their time doing laborious and repetitive tasks, be it pipetting liquid samples or running the same analyses over and over again. But what if they could simply tell a robot to do the experiments, analyze the data, and generate a report? 

Enter Organa, a benchtop robotic system devised by researchers at the University of Toronto that can perform chemistry experiments. In a paper posted on the arXiv preprint server, the team reported that the system could automate some chemistry lab tasks using a combination of computer vision and a large language model (LLM) that translates scientists’ verbal cues into an experimental pipeline. 

Imagine having a robot that can collaborate with a human scientist on a chemistry experiment, says Alán Aspuru-Guzik, a chemist, computer scientist, and materials scientist at the University of Toronto, who is one of the project’s leaders. Aspuru-Guzik’s vision is to elevate traditional lab automation to “eventually make an AI scientist,” one that can perform and troubleshoot an experiment and even offer feedback on the results. 

Aspuru-Guzik and his team designed Organa to be flexible. That means that instead of performing only one task or one part of an experiment as a typical fixed automation system would, it can perform a multistep experiment on cue. The system is also equipped with visualization tools that can monitor progress and provide feedback on how the experiment is going.  

“This is one of the early examples of showing how you can have a bidirectional conversation with an AI assistant for a robotic chemistry lab,” says Milad Abolhasani, a chemical and material engineer at North Carolina State University, who was not involved in the project. 

Most automated lab equipment is not easily customizable or reprogrammable to suit the chemists’ needs, says Florian Shkurti, a computer scientist at the University of Toronto and a co-leader of the project. And even if it is, the chemists would need to have programming skills. But with Organa, scientists can simply convey their experiments through speech. As scientists prompt the robot with their experimental objectives and setup, Organa’s LLM translates this natural-language instruction into χDL codes, a standard chemical description language. The algorithm breaks down the codes into steps and goals, with a road map to execute each task. If there is an ambiguous instruction or an unexpected outcome, it can flag the issue for the scientist to resolve.

About two-thirds of Organa’s hardware components are made from off-the-shelf parts, making it easier to replicate across laboratories, Aspuru-Guzik says. The robot has a camera detector that can identify both opaque objects and transparent ones, such as a chemical flask.

Organa’s first task was to characterize the electrochemical properties of quinones, the electroactive molecules used in rechargeable batteries. The experiment has 19 parallel steps, including routine chemistry steps such as pH and solubility tests, recrystallization, and an electrochemical measurement. It also involves a tedious electrode-precleaning step, which takes up to six hours. “Chemists really, really hate this,” says Shkurti.

Organa completed the 19-step experiment in about the same amount of time it would take a human—and with comparable results. While the efficiency was not noticeably better than in a manual run, the robot can be much more productive if it is run overnight. “We always get the advantage of it being able to work 24 hours,” Shkurti says. Abolhasani adds, “That’s going to save a lot of our highly trained scientists time that they can use to focus on thinking about the scientific problem, not doing these routine tasks in the lab.” 

Organa’s most sophisticated feature is perhaps its ability to provide feedback on generated data. “We were surprised to find that this visual language model can spot outliers on chemistry graphs,” explains Shkurti. The system also flags these ambiguities or uncertainties and suggests methods of troubleshooting. 

The group is now working on improving the LLM’s ability to plan tasks and then revise those plans to make the system more amenable to experimental uncertainties. 

“There’s a lot roboticists have to offer to scientists in order to amplify what they can do and get them better data,” Shkurti says. “I am really excited to try to create new possibilities.” 

Kristel Tjandra is a freelance science writer based in Oahu. 

In an age where customer experience can make or break a business, Cathay Pacific is embracing cloud transformation to enhance service delivery and revolutionize operations from the inside out. It’s not just technology companies that are facing pressure to deliver better customer service, do more with data, and improve agility. An almost 80-year-old airline, Cathay Pacific embarked on its digital transformation journey in 2014, spurred by a critical IT disruption that became the catalyst for revamping their technology.

By embracing the cloud, the airline has not only streamlined operations but also paved the way for innovative solutions like DevSecOps and AI integration. This shift has enabled Cathay to deliver faster, more reliable services to both passengers and staff, while maintaining a robust security framework in an increasingly digital world. 

According to Rajeev Nair, general manager of IT infrastructure and security at Cathay Pacific, becoming a digital-first airline was met with early resistance from both business and technical teams. The early stages required a lot of heavy lifting as they shifted legacy apps first from their server room to a dedicated data center and then to the cloud. From there began the process of modernization that Cathay Pacific, now in its final stages of this transformation, continues to fine tune.

The cloud migration also helped Cathay align with their ESG goals. “Two years ago, if you asked me what IT could do for sustainability, we would’ve been clueless,” says Nair. However, through cloud-first strategies and green IT practices, the airline has made notable strides in reducing its carbon footprint. Currently, the business is in the process of moving to a smaller data center, reducing physical infrastructure and its carbon emissions significantly by 2025.

The broader benefits of this cloud transformation for Cathay Pacific go beyond sustainability. Agility, time-to-market, and operational efficiency have improved drastically. “If you ask many of the enterprises, they would probably say that shifting to the cloud is all about cost-saving,” says Nair. “But for me, those are secondary aspects and the key is about how to enable the business to be more agile and nimble so that the business capability could be delivered much faster by IT and the technology team.”

By 2025, Cathay Pacific aims to have 100% of their business applications running on the cloud, significantly enhancing their agility, customer service, and cost efficiency, says Nair.

As Cathay Pacific continues its digital evolution, Nair remains focused on future-proofing the airline through emerging technologies. Looking ahead, he is particularly excited about the potential of AI, generative AI, and virtual reality to further enhance both customer experience and internal operations. From more immersive VR-based training for cabin crew to enabling passengers to preview in-flight products before boarding, these innovations are set to redefine how the airline engages with its customers and staff. 

“We have been exploring that for quite some time, but we believe that it will continue to be a mainstream technology that can change the way we serve the customer,” says Nair.

This episode of Business Lab is produced in association with Infosys Cobalt.

Full Transcript 

Megan Tatum: From MIT Technology Review, I’m Megan Tatum, and this is Business Lab, the show that helps business leaders make sense of new technologies coming out of the lab and into the marketplace. 

Our topic today is cloud transformation to meet business goals and customer needs. It’s not just tech companies that have to stay one step ahead. Airlines too are under pressure to deliver better customer service, do more with data, and improve agility. 

Two words for you: going further. 

My guest is Rajeev Nair, who is the general manager of IT infrastructure and security at Cathay Pacific. This podcast is produced in association with Infosys Cobalt. Welcome, Rajeev. 

Rajeev Nair: Thank you. Thank you, Megan. Thank you for having me. 

Megan: Thank you ever so much for joining us. Now to get some context for our conversation today, could you first describe how Cathay Pacific’s digital transformation journey began, and explain, I guess, what stage of this transformation this almost 80-year-old airline is currently in, too? 

Rajeev: Sure, definitely Megan. So for Cathay, we started this transformation journey probably a decade back, way back in 2014. It all started with facing some major service disruption within Cathay IT where it had a massive impact on the business operation. That prompted us to trigger and initiate this transformation journey. So the first thing is we started looking at many of our legacy applications. Back in those days we still had mainframe systems that provided so many of our critical services. We started looking at migrating those legacy apps first, moving them outside of that legacy software and moving them into a proper data center. Back in those days, our data center used to be our corporate headquarters. We didn’t have a dedicated data center and it used to be in a server room. So those were the initial stages of our transformation journey, just a basic building block. So we started moving into a proper data center so that resilience and availability could be improved. 

And as a second phase, we started looking at the cloud. Those days, cloud was just about to kick off in this part of the world. We started looking at migrating to the cloud and it has been a huge challenge or resistance even from the business as well as from the technology team. Once we started moving, shifting apps to the cloud, we had multiple transformation programs to do that modernization activities. Once that is done, then the third phase of the journey is more about your network. Once your applications are moved to the cloud, your network design needs to be completely changed. Then we started looking at how we could modernize our network because Cathay operates in about 180 regions across the world. So our network is very crucial for us. We started looking at redesigning our network. 

And then, it comes to your security aspects. Things moving to the cloud, your network design is getting changed, your cybersecurity needs heavy lifting to accommodate the modern world. We started focusing on cybersecurity initiatives where our security posture has been improved a lot over the last few years. And with those basic building blocks done on the hardware and on the technology side, then comes your IT operations. Because one is your hardware and software piece, but how do you sustain your processes to ensure that it can support those changing technology landscapes? We started investing a lot around the IT operations side, but things like ITIL processes have been revisited. We started adopting many of the DevOps and the DevSecOps practices. So a lot of emphasis around processes and practices to help the team move forward, right? 

And those operations initiatives are in phase. As we stand today, we are at the final stage of our cloud journey where we are looking at how we can optimize it better. So we shifted things to the cloud and that has been a heavy lifting that has been done in the early phases. Now we are focusing around how we can rewrite or refactor your application so that it can better liberate your cloud technologies where we could optimize the performance, thereby optimizing your usage and the cloud resources wherein you could save on the cost as well as on the sustainability aspect. That is where we stand. By 2025, we are looking at moving 100% of our business applications to the cloud and also reducing our physical footprint in our data centers as well. 

Megan: Fantastic. And you mentioned sustainability there. I wonder how does the focus on environmental, social, and governance goals or ESG tie into your wider technology strategy? 

Rajeev: Sure. And to be very honest, Megan, if you asked me this question two years back, we would’ve been clueless on what IT could do from a sustainability aspect. But over the last two years, there has been a lot of focus around ESG components within the technology space where we have done a lot of initiatives since last year to improve and be efficient on the sustainability front. So a couple of key areas that we have done. One is definitely the cloud-first strategy where adopting the cloud-first policy reduces your carbon footprint and it also helps us in migrating away from our data center. So as we speak, we are doing a major project to further reduce our data center size by relocating to a much smaller data center, which will be completed by the end of next year. That will definitely help us to reduce our footprint. 

The second is around adopting the various green IT practices, things like energy efficient devices, be it your PCs or the laptop or virtualizations, and e-based management policies and management aspects. Some of the things are very basic and fundamental in nature. Stuff like we moved away from a dual monitor to a single monitor wherein we could reduce your energy consumption by half, or changing some of your software policies like screen timeouts and putting a monitor in standby. Those kinds of basic things really helped us to optimize and manage. And the last one is around FinOps. So FinOps is a process in the practice that is being heavily adopted in the cloud organization, but it is just not about optimizing your course because by adopting the FinOps practices and tying in with the GreenOps processes, we are able to focus a lot around reducing our CO2 footprint and optimizing sustainability. Those are some of the practices that we have been doing with Cathay. 

Megan: Yeah. fantastic benefits from relatively small changes there. Other than ESG, what are the other benefits for an enterprise like Cathay Pacific in terms of shifting from those legacy systems to the cloud that you found? 

Rajeev: For me, the key is about agility and time-to-market capability. If you ask many of the enterprises, they would probably say that shifting to the cloud is all about cost-saving. But for me, those are secondary aspects. The key is about how to enable the business to be more agile and nimble so that the business capability can be delivered much faster by IT and the technology team. So as an example, gone are the days when we take about a few months before we provision hardware and have the platform and the applications ready. Now the platforms are being delivered to the developers within an hour’s time so that the developers can quickly build their development environment and be ready for development and testing activities. Right? So agility is a key and the number one factor. 

The second is by shifting to the cloud, you’re also liberating many of the latest technologies that the cloud comes up with and the provider has to offer. Things like capacity and the ability to scale up and down your resources and services according to your business needs and fluctuations are a huge help from a technology aspect. That way you can deliver customer-centered solutions faster and more efficiently than many of our airline customers and competitors. 

And the last one is, of course, your cost saving aspect and the operational efficiency. By moving away from the legacy systems, we can reduce a lot of capex [capital expenditure]. Like, say for example, I don’t need to spend money on investing in hardware and spend resources to manage those hardware and data center operations, especially in Hong Kong where human resources are pretty expensive and scarce to find. It is very important that I rely on these sorts of technologies to manage those optimally. Those are some of the key aspects that we see from a cloud adoption perspective. 

Megan: Fantastic. And it sounds like it’s been a several year process so far. So after what sounds like pretty heavy investment when it comes to moving legacy hardware on-prem systems to the cloud. What’s your approach now to adapting your IT operations off the back of that? 

Rajeev: Exactly. That is, sort of, just based early in my transformation journey, but yeah, absolutely. By moving to the cloud, it is just not about the hardware, but it’s also about how your operations and your processes align with this changing technology and new capabilities. And, for example, by adopting more agile and scalable approach to managing IT infrastructures and applications as well. Also leveraging the data and insights that the cloud enables. To achieve this, the fundamental aspect of this is how you can revisit and fine tune your IT service management processes, and that is where your core of IT operations have been built in the past. And to manage that properly we recently, I think, over the last three years we were looking at implementing a new IT service management solution, which is built on a product called ServiceNow. So they are built on the core ITIL processes framework to help us manage the service management, the operations management, and asset management. 

Those are some of the capabilities which we rolled out with the help of our partners like Infosys so that it could provide a framework to fine tune and optimize IT processes. And we also adopted things like DevOps and DevSecOps because what we have also noticed is the processes like ITIL, which was very heavy over the last few years around support activities is also shifting. So we wanted to adopt some of these development practices into the support and operations functions to be more agile by shifting left some of these capabilities. And in this journey, Infosys has been our key partner, not only on the cloud transformation side, but also on implementation of ServiceNow, which is our key service management tool where they provided us end-to-end support starting from the planning phase or the initial conceptual phase and also into the design and development and also to the deployment and maintenance. We haven’t completed this journey and it’s still a project that is currently ongoing, and by 2025 we should be able to complete this successfully across the enterprise. 

Megan: Fascinating. It’s an awful lot of change going on. I mean, there must be an internal shift, therefore, that comes with cloud transformation too, I imagine. I wonder, what’s your approach been to up skilling your team to help it excel in this new way of working? 

Rajeev: Yeah, absolutely. And that is always the hardest part. You can change your technology and processes is but changing your people, that’s always toughest and the hardest bit. And essentially this is all about change management, and that has been one of our struggles in our early part of the cloud transformation journey. What we did is we invested a lot in terms of uplifting our traditional infrastructure team. All the traditional technology teams have to go through that learning curve in adopting cloud technology early in our project. And we also provided a lot of training programs, including some of our cloud partners were able to up skill and train these resources. 

But the key differences that we are seeing is even after providing all those training and upskilling programs, we could see that there was a lot of resistance and a lot of doubts in people’s mind about how cloud is going to help the organization. And the best part is what we did is we included these team members into our project so that they get the hands-on experience. And once they start seeing the benefits around these technologies, there was no looking back. And the team was able to completely embrace the cloud technologies to the point that we still have a traditional technology team who’s supporting the remaining hardware and the servers of the world, but they’re also very keen to shift across the line and adopt and embrace the cloud technology. But it’s been quite a journey for us. 

Megan: That’s great to hear that you’ve managed to bring them along with you. And I suppose it’d be remiss of me if we’re talking about embracing new technologies not to talk about AI, although still in its early stages in most industries. I wonder how is Cathay Pacific approaching AI adoption as well? 

Rajeev: Sure. I think these days none of these conversations can be complete without talking about AI and gen AI. We started this early exploratory phase early into the game, especially in this part of the world. But for us, the key is approaching this based on the customer’s pain points and business needs and then we work backward to identify what type of AI is best suitable or relevant to us. In Cathay, currently, we focus on three main types of AI. One is of course conversational AI. Essentially, it is a form of an internal and external chatbot. Our chatbot, we call it Vera, serves customers directly and can handle about 50% of the inquiries successfully. And just about two weeks back, we upgraded the LLM with a new model, the chatbot with a new model, which is able to be more efficient and much more responsive in terms of the human work. So that’s one part of the AI that we heavily invested on. 

Second is RPA, or robotic process automation, especially what you’re seeing is during the pandemic and post-Covid era, there is limited resources available, especially in Hong Kong and across our supply chain. So RPA or the robotic processes helps to automate mundane repetitive tasks, which doesn’t only fill the resource gap, but it also directly enhances the employee experience. And so far in Cathay, we have about a hundred bots in production serving various business units, serving approximately 30,000 hours every year of human activity. So that’s the second part. 

The third one is around ML and it’s the gen AI. So like our digital team or the data science team has developed about 70-plus ML models in Cathay that turned the organization data into insights or actionable items. These models help us to make a better decision. For example, what meals to be loaded into the aircraft and specific routes, in terms of what quantity and what kind of product offers we promote to customers, and including the fare loading and the pricing of our passenger as well as a cargo bay space. There is a lot of exploration that is being done in this space as well. And a couple of examples I could relate is if you ever happen to come to Hong Kong, next time at the airport, you could hear the public announcement system and that is also AI-powered recently. In the past, our staff used to manually make those announcements and now it has been moved away and has been moved into AI-powered voice technology so that we could be consistent in our announcement. 

Megan: Oh, fantastic. I’ll have to listen for it next time I’m at Hong Kong airport. And you’ve mentioned this topic a couple of times in the conversation. Look, when we’re talking about cloud modernization, cybersecurity can be a roadblock to agility, I guess, if it’s not managed effectively. So could you also tell us in a little more detail how Cathay Pacific has integrated security into its digital transformation journey, particularly with the adoption of development security operations practices that you’ve mentioned? 

Rajeev: Yeah, this is an interesting one. I look after cybersecurity as well as the infrastructure services. With both of these critical functions around my hand, I need to be mindful of both aspects, right? Yes, it’s an interesting one and it has changed over the period of time, and I fully understand why cybersecurity practices needs to be rigid because there is a lot of compliance and it is a highly regulated function, but if something goes wrong, as a CISO we are held accountable for those faults. I can understand why the team is so rigid in their practices. And I also understand from a business perspective it could be perceived as a road blocker to agility. 

One of the key aspects that we have done in Cathay is we have been following DevOps for quite a number of years, and recently, I think in the last two years, we started implementing DevSecOps into our STLC [software testing life cycle]. And what it essentially means is rather than the core cybersecurity team being responsible for many of the security testing and those sorts of aspects, we want to shift left some of these capabilities into the developers so that the people who develop the code now are held accountable for the testing and the quality of the output. And they’re also enabled in terms of the cybersecurity process. Right? 

Of course, when we started off this journey, there has been a huge resistance on the security team itself because they don’t really trust the developers trying to do the testing or the testing outputs. But over a period of time with the introduction of various tools and automation that is put in place, this is now getting into a matured stage wherein it is now enabling the upfront teams to take care of all the aspects of security, like threat modeling, code scanning, and the vulnerability testing. But at the end, the security teams would be still validating and act as a sort of a gatekeeper, but in a very light and inbuilt processes. And this way we can ensure that our cloud applications are secure by design and by default they can deliver them faster and more reliably to our customers. And in this entire process, right? 

In the past, security has been always perceived as an accountability of the cybersecurity team. And by enabling the developers of the security aspects, now you have a better ownership in the organization when it comes to cybersecurity and it is building a better cybersecurity culture within the organization. And that, to me, is a key because from a security aspect, we always say that people are your first line of defense and often they’re also the last line of defense. I’m glad that by these processes we are able to improve that maturity in the organization. 

Megan: Absolutely. And you mentioned that obviously cybersecurity is something that’s really important to a lot of customers nowadays as well. I wondered if you could offer some other examples too of how your digital transformation has improved that customer experience in other ways? 

Rajeev: Yeah, definitely. Maybe I can quote a few examples, Megan. One is around our pilots. You would’ve seen when you travel through the airport or in the aircraft that pilots usually carry a briefcase when they load the flight, and you are often probably wondering what exactly they carry. Basically, that contains a bunch of papers. It contains your weather charts, your navigation routes, and the flight plans, the crew details. It’s a whole stack of paper that they have to carry on each and every flight. And in Cathay, by digitization, we have automated that in their processes, where now they carry an iPad instead of a bunch of papers or briefing pack. So that iPad includes all these softwares that is required for the captain to operate the flight in a legally and a safe manner. 

Paperless cockpit operation is nothing new. Many airlines have attempted to do that, but I should say that Cathay has been on the forefront in truly establishing a paperless operation, where many of the other airlines have shown great interest in using our software. That is one aspect from a fly crew perspective. Second, from a customer perspective, we have an app called Customer 360, which is a completely in-house developed model, which has all the customer direct transactions, surveys, or how they interact at the various checkpoints with our crew or at the boarding. You have all this data feed of a particular customer where our agents or the cabin crew can understand the customer’s sentiment and their reaction to service recovery action. 

Say for example, the customer calls up a call center and ask for a refund or miles compensation. Based on the historical usage, we could prioritize the best action to improve the customer satisfaction. We are connected to all these models and enable the frontline teams so that they can use this when they engage with the customer. An example at the airport, our agents will be able to see a lot of useful insights about the customers beyond the basic information like the flight itinerary or the online shopping history at the Cathay shop, et cetera, so that they can see the overall satisfaction level and get additional insights on recommended actions to restore or improve the customer satisfaction level. This is basically used by our frontline agents at the airport, our cabin crew as well as all the airport team, and the customer team so that they have great consistency in the service no matter what touchpoint the customers are choosing to contact us. 

Megan: Fantastic. 

Rajeev: So these are a few example looking from a back end as well as from a front line of the team perspective. 

Megan: Yeah, absolutely. I’m sure there’s a few people listening who were wondering what pilots carry in that suitcase. So thank you so much for clearing that up. And finally, Rajeev, I guess looking ahead, what emerging technologies are you excited to explore further going forward to enhance digital capabilities and customer experience in the years to come? 

Rajeev: Yeah, so we will continue to explore AI and gen AI capability, which has been the spotlight for the last 18 months or so, be it for the passenger or even for the staff internally. We will continue to explore that. But apart from AI, one other aspect I believe could go at great ways around the AR and the VR capabilities, basically virtual reality. We have been exploring that for quite some time, but we believe that it will continue to be a mainstream technology that can change the way we serve the customer. Say for example, in Cathay, we already have a VR cave for our cabin crew training, virtual reality capabilities, and in a few months’ time, we are actually launching a learning facility based on VR where we could be able to provide more immersive learning experience for the cabin crew and later for the other employees. 

Basically, before a cabin crew is able to operate a flight, they go through a rigorous training in Cathay City in our headquarters, basically to know how to serve our passengers, how to handle an emergency situation, those sorts of aspects. And in many cases, we travel the crew from various outports or various countries back into Hong Kong to train them and equip them for these training activities. You can imagine that costs us a lot of money and effort to bring all the people back to Hong Kong. And by having VR capabilities, we are able to do that anywhere in the world without having that physical presence. That’s one area where it’ll go mainstream. 

The second is around other business units. Apart from the cabin crew, we are also experimenting the VR on the customer front. For example, we are able to launch a new business class seat product we call the Aria Suite by next year. And VR technology will help the customers to visualize the seat details without them able to get on board. So without them flying, even before that, they’re able to experience a product on the ground. At our physical shop in Hong Kong, customers can now use a virtual reality technology to visualize how our designer furniture and lifestyle products fit in the sitting rooms. The list of VR capabilities goes very long. The list goes on. And this is also a great and important way to engage with our customers in particular. 

Megan: Wow. Sounds like some exciting stuff on the way. Thank you ever so much, Rajeev, for talking us through that. That was Rajeev Nair, the general manager of IT infrastructure and security at Cathay Pacific, who I spoke with from an unexpectedly sunny Brighton, England.

That’s it for this episode of Business Lab. I’m your host, Megan Tatum. I’m a contributing editor and host for Insights, the custom publishing division of MIT Technology Review. We were founded in 1899 at the Massachusetts Institute of Technology, and you can find us in print, on the web, and at events each year around the world. For more information about us and the show, please check out our website at technologyreview.com. 

This show is available wherever you get your podcasts. If you enjoyed this episode, we hope you’ll take a moment to rate and review us. Business Lab is a production of MIT Technology Review, this episode was produced by Giro Studios. Thanks for listening. 

Big-name makers of processors, especially those geared toward cloud-based AI, such as AMD and Nvidia, have been showing signs of wanting to own more of the business of computing, purchasing makers of software, interconnects, and servers. The hope is that control of the “full stack” will give them an edge in designing what their customers want.

Amazon Web Services (AWS) got there ahead of most of the competition, when they purchased chip designer Annapurna Labs in 2015 and proceeded to design CPUs, AI accelerators, servers, and data centers as a vertically-integrated operation. Ali Saidi, the technical lead for the Graviton series of CPUs, and Rami Sinno, director of engineering at Annapurna Labs, explained the advantage of vertically-integrated design and Amazon-scale and showed IEEE Spectrum around the company’s hardware testing labs in Austin, Tex., on 27 August.

Saidi and Sinno on:

• What kind of engineer a vertically-integrated cloud company needs
• How the Graviton series of CPUs evolved
• How chip design is different at AWS
• How the lab location speeds design

What brought you to Amazon Web Services, Rami?

Rami SinnoAWS

Rami Sinno: Amazon is my first vertically integrated company. And that was on purpose. I was working at Arm, and I was looking for the next adventure, looking at where the industry is heading and what I want my legacy to be. I looked at two things:

One is vertically integrated companies, because this is where most of the innovation is—the interesting stuff is happening when you control the full hardware and software stack and deliver directly to customers.

And the second thing is, I realized that machine learning, AI in general, is going to be very, very big. I didn’t know exactly which direction it was going to take, but I knew that there is something that is going to be generational, and I wanted to be part of that. I already had that experience prior when I was part of the group that was building the chips that go into the Blackberries; that was a fundamental shift in the industry. That feeling was incredible, to be part of something so big, so fundamental. And I thought, “Okay, I have another chance to be part of something fundamental.”

Does working at a vertically-integrated company require a different kind of chip design engineer?

Sinno: Absolutely. When I hire people, the interview process is going after people that have that mindset. Let me give you a specific example: Say I need a signal integrity engineer. (Signal integrity makes sure a signal going from point A to point B, wherever it is in the system, makes it there correctly.) Typically, you hire signal integrity engineers that have a lot of experience in analysis for signal integrity, that understand layout impacts, can do measurements in the lab. Well, this is not sufficient for our group, because we want our signal integrity engineers also to be coders. We want them to be able to take a workload or a test that will run at the system level and be able to modify it or build a new one from scratch in order to look at the signal integrity impact at the system level under workload. This is where being trained to be flexible, to think outside of the little box has paid off huge dividends in the way that we do development and the way we serve our customers.

“By the time that we get the silicon back, the software’s done” —Ali Saidi, Annapurna Labs

At the end of the day, our responsibility is to deliver complete servers in the data center directly for our customers. And if you think from that perspective, you’ll be able to optimize and innovate across the full stack. A design engineer or a test engineer should be able to look at the full picture because that’s his or her job, deliver the complete server to the data center and look where best to do optimization. It might not be at the transistor level or at the substrate level or at the board level. It could be something completely different. It could be purely software. And having that knowledge, having that visibility, will allow the engineers to be significantly more productive and delivery to the customer significantly faster. We’re not going to bang our head against the wall to optimize the transistor where three lines of code downstream will solve these problems, right?

Do you feel like people are trained in that way these days?

Sinno: We’ve had very good luck with recent college grads. Recent college grads, especially the past couple of years, have been absolutely phenomenal. I’m very, very pleased with the way that the education system is graduating the engineers and the computer scientists that are interested in the type of jobs that we have for them.

The other place that we have been super successful in finding the right people is at startups. They know what it takes, because at a startup, by definition, you have to do so many different things. People who’ve done startups before completely understand the culture and the mindset that we have at Amazon.

[back to top]

What brought you to AWS, Ali?

Ali SaidiAWS

Ali Saidi: I’ve been here about seven and a half years. When I joined AWS, I joined a secret project at the time. I was told: “We’re going to build some Arm servers. Tell no one.”

We started with Graviton 1. Graviton 1 was really the vehicle for us to prove that we could offer the same experience in AWS with a different architecture.

The cloud gave us an ability for a customer to try it in a very low-cost, low barrier of entry way and say, “Does it work for my workload?” So Graviton 1 was really just the vehicle demonstrate that we could do this, and to start signaling to the world that we want software around ARM servers to grow and that they’re going to be more relevant.

Graviton 2—announced in 2019—was kind of our first… what we think is a market-leading device that’s targeting general-purpose workloads, web servers, and those types of things.

It’s done very well. We have people running databases, web servers, key-value stores, lots of applications... When customers adopt Graviton, they bring one workload, and they see the benefits of bringing that one workload. And then the next question they ask is, “Well, I want to bring some more workloads. What should I bring?” There were some where it wasn’t powerful enough effectively, particularly around things like media encoding, taking videos and encoding them or re-encoding them or encoding them to multiple streams. It’s a very math-heavy operation and required more [single-instruction multiple data] bandwidth. We need cores that could do more math.

We also wanted to enable the [high-performance computing] market. So we have an instance type called HPC 7G where we’ve got customers like Formula One. They do computational fluid dynamics of how this car is going to disturb the air and how that affects following cars. It’s really just expanding the portfolio of applications. We did the same thing when we went to Graviton 4, which has 96 cores versus Graviton 3’s 64.

[back to top]

How do you know what to improve from one generation to the next?

Saidi: Far and wide, most customers find great success when they adopt Graviton. Occasionally, they see performance that isn’t the same level as their other migrations. They might say “I moved these three apps, and I got 20 percent higher performance; that’s great. But I moved this app over here, and I didn’t get any performance improvement. Why?” It’s really great to see the 20 percent. But for me, in the kind of weird way I am, the 0 percent is actually more interesting, because it gives us something to go and explore with them.

Most of our customers are very open to those kinds of engagements. So we can understand what their application is and build some kind of proxy for it. Or if it’s an internal workload, then we could just use the original software. And then we can use that to kind of close the loop and work on what the next generation of Graviton will have and how we’re going to enable better performance there.

What’s different about designing chips at AWS?

Saidi: In chip design, there are many different competing optimization points. You have all of these conflicting requirements, you have cost, you have scheduling, you’ve got power consumption, you’ve got size, what DRAM technologies are available and when you’re going to intersect them… It ends up being this fun, multifaceted optimization problem to figure out what’s the best thing that you can build in a timeframe. And you need to get it right.

One thing that we’ve done very well is taken our initial silicon to production.

How?

Saidi: This might sound weird, but I’ve seen other places where the software and the hardware people effectively don’t talk. The hardware and software people in Annapurna and AWS work together from day one. The software people are writing the software that will ultimately be the production software and firmware while the hardware is being developed in cooperation with the hardware engineers. By working together, we’re closing that iteration loop. When you are carrying the piece of hardware over to the software engineer’s desk your iteration loop is years and years. Here, we are iterating constantly. We’re running virtual machines in our emulators before we have the silicon ready. We are taking an emulation of [a complete system] and running most of the software we’re going to run.

So by the time that we get to the silicon back [from the foundry], the software’s done. And we’ve seen most of the software work at this point. So we have very high confidence that it’s going to work.

The other piece of it, I think, is just being absolutely laser-focused on what we are going to deliver. You get a lot of ideas, but your design resources are approximately fixed. No matter how many ideas I put in the bucket, I’m not going to be able to hire that many more people, and my budget’s probably fixed. So every idea I throw in the bucket is going to use some resources. And if that feature isn’t really important to the success of the project, I’m risking the rest of the project. And I think that’s a mistake that people frequently make.

Are those decisions easier in a vertically integrated situation?

Saidi: Certainly. We know we’re going to build a motherboard and a server and put it in a rack, and we know what that looks like… So we know the features we need. We’re not trying to build a superset product that could allow us to go into multiple markets. We’re laser-focused into one.

What else is unique about the AWS chip design environment?

Saidi: One thing that’s very interesting for AWS is that we’re the cloud and we’re also developing these chips in the cloud. We were the first company to really push on running [electronic design automation (EDA)] in the cloud. We changed the model from “I’ve got 80 servers and this is what I use for EDA” to “Today, I have 80 servers. If I want, tomorrow I can have 300. The next day, I can have 1,000.”

We can compress some of the time by varying the resources that we use. At the beginning of the project, we don’t need as many resources. We can turn a lot of stuff off and not pay for it effectively. As we get to the end of the project, now we need many more resources. And instead of saying, “Well, I can’t iterate this fast, because I’ve got this one machine, and it’s busy.” I can change that and instead say, “Well, I don’t want one machine; I’ll have 10 machines today.”

Instead of my iteration cycle being two days for a big design like this, instead of being even one day, with these 10 machines I can bring it down to three or four hours. That’s huge.

How important is Amazon.com as a customer?

Saidi: They have a wealth of workloads, and we obviously are the same company, so we have access to some of those workloads in ways that with third parties, we don’t. But we also have very close relationships with other external customers.

So last Prime Day, we said that 2,600 Amazon.com services were running on Graviton processors. This Prime Day, that number more than doubled to 5,800 services running on Graviton. And the retail side of Amazon used over 250,000 Graviton CPUs in support of the retail website and the services around that for Prime Day.

[back to top]

The AI accelerator team is colocated with the labs that test everything from chips through racks of servers. Why?

Sinno: So Annapurna Labs has multiple labs in multiple locations as well. This location here is in Austin… is one of the smaller labs. But what’s so interesting about the lab here in Austin is that you have all of the hardware and many software development engineers for machine learning servers and for Trainium and Inferentia [AWS’s AI chips] effectively co-located on this floor. For hardware developers, engineers, having the labs co-located on the same floor has been very, very effective. It speeds execution and iteration for delivery to the customers. This lab is set up to be self-sufficient with anything that we need to do, at the chip level, at the server level, at the board level. Because again, as I convey to our teams, our job is not the chip; our job is not the board; our job is the full server to the customer.

How does vertical integration help you design and test chips for data-center-scale deployment?

Sinno: It’s relatively easy to create a bar-raising server. Something that’s very high-performance, very low-power. If we create 10 of them, 100 of them, maybe 1,000 of them, it’s easy. You can cherry pick this, you can fix this, you can fix that. But the scale that the AWS is at is significantly higher. We need to train models that require 100,000 of these chips. 100,000! And for training, it’s not run in five minutes. It’s run in hours or days or weeks even. Those 100,000 chips have to be up for the duration. Everything that we do here is to get to that point.

We start from a “what are all the things that can go wrong?” mindset. And we implement all the things that we know. But when you were talking about cloud scale, there are always things that you have not thought of that come up. These are the 0.001-percent type issues.

In this case, we do the debug first in the fleet. And in certain cases, we have to do debugs in the lab to find the root cause. And if we can fix it immediately, we fix it immediately. Being vertically integrated, in many cases we can do a software fix for it. We use our agility to rush a fix while at the same time making sure that the next generation has it already figured out from the get go.

[back to top]

A team in Australia has recently demonstrated a key advance in metal-oxide-semiconductor-based (or MOS-based) quantum computers. They showed that their two-qubit gates—logical operations that involve more than one quantum bit, or qubit—perform without errors 99 percent of the time. This number is important, because it is the baseline necessary to perform error correction, which is believed to be necessary to build a large-scale quantum computer. What’s more, these MOS-based quantum computers are compatible with existing CMOS technology, which will make it more straightforward to manufacture a large number of qubits on a single chip than with other techniques.

“Getting over 99 percent is significant because that is considered by many to be the error correction threshold, in the sense that if your fidelity is lower than 99 percent, it doesn’t really matter what you’re going to do in error correction,” says Yuval Boger, CCO of quantum computing company QuEra and who wasn’t involved in the work. “You’re never going to fix errors faster than they accumulate.”

There are many contending platforms in the race to build a useful quantum computer. IBM, Google and others are building their machines out of superconducting qubits. Quantinuum and IonQ use individual trapped ions. QuEra and Atom Computing use neutrally-charged atoms. Xanadu and PsiQuantum are betting on photons. The list goes on.

In the new result, a collaboration between the University of New South Wales (UNSW) and Sydney-based startup Diraq, with contributors from Japan, Germany, Canada, and the U.S., has taken yet another approach: trapping single electrons in MOS devices. “What we are trying to do is we are trying to make qubits that are as close to traditional transistors as they can be,” says Tuomo Tanttu, a research fellow at UNSW who led the effort.

Qubits That Act Like Transistors

These qubits are indeed very similar to a regular transistor, gated in such a way as to have only a single electron in the channel. The biggest advantage of this approach is that it can be manufactured using traditional CMOS technologies, making it theoretically possible to scale to millions of qubits on a single chip. Another advantage is that MOS qubits can be integrated on-chip with standard transistors for simplified input, output, and control, says Diraq CEO Andrew Dzurak.

The drawback of this approach, however, is that MOS qubits have historically suffered from device-to-device variability, causing significant noise on the qubits.

“The sensitivity in [MOS] qubits is going to be more than in transistors, because in transistors, you still have 20, 30, 40 electrons carrying the current. In a qubit device, you’re really down to a single electron,” says Ravi Pillarisetty, a senior device engineer for Intel quantum hardware who wasn’t involved in the work.

The team’s result not only demonstrated the 99 percent accurate functionality on two-qubit gates of the test devices, but also helped better understand the sources of device-to-device variability. The team tested three devices with three qubits each. In addition to measuring the error rate, they also performed comprehensive studies to glean the underlying physical mechanisms that contribute to noise.

The researchers found that one of the sources of noise was isotopic impurities in the silicon layer, which, when controlled, greatly reduced the circuit complexity necessary to run the device. The next leading cause of noise was small variations in electric fields, likely due to imperfections in the oxide layer of the device. Tanttu says this is likely to improve by transitioning from a laboratory clean room to a foundry environment.

“It’s a great result and great progress. And I think it’s setting the right direction for the community in terms of thinking less about one individual device, or demonstrating something on an individual device, versus thinking more longer term about the scaling path,” Pillarisetty says.

Now, the challenge will be to scale up these devices to more qubits. One difficulty with scaling is the number of input/output channels required. The quantum team at Intel, who are pursuing a similar technology, has recently pioneered a chip they call Pando Tree to try to address this issue. Pando Tree will be on the same plane as the quantum processor, enabling faster inputs and outputs to the qubits. The Intel team hopes to use it to scale to thousands of qubits. “A lot of our approach is thinking about, how do we make our qubit processor look more like a modern CPU?” says Pillarisetty.

Similarly, Diraq CEO Dzurak says his team plan to scale their technology to thousands of qubits in the near future through a recently announced partnership with Global Foundries. “With Global Foundries, we designed a chip that will have thousands of these [MOS qubits]. And these will be interconnected by using classical transistor circuitry that we designed. This is unprecedented in the quantum computing world,” Dzurak says.

It pays to have friends in fascinating places. You need look no further than the cover of this issue and the article “ IBM’s Big Bet on the Quantum-Centric Supercomputer” for evidence. The article by Ryan Mandelbaum, Antonio D. Córcoles, and Jay Gambetta came to us courtesy of the article’s illustrator, the inimitable graphic artist Carl De Torres, a longtime IEEE Spectrum contributor as well as a design and communications consultant for IBM Research.

Story ideas typically originate with Spectrum’s editors and pitches from expert authors and freelance journalists. So we were intrigued when De Torres approached Spectrum about doing an article on IBM Research’s cutting-edge work on quantum-centric supercomputing.

De Torres has been collaborating with IBM in a variety of capacities since 2009, when, while at Wired magazine creating infographics, he was asked by the ad agency Ogilvy to work on Big Blue’s advertising campaign “Let’s build a Smarter Planet.” That project went so well that De Torres struck out on his own the next year. His relationship with IBM expanded, as did his engagements with other media, such as Spectrum, Fortune, and The New York Times. “My interest in IBM quickly grew beyond helping them in a marketing capacity,” says De Torres, who owns and leads the design studio Optics Lab in Berkeley, Calif. “What I really wanted to do is get to the source of some of the smartest work happening in technology, and that was IBM Research.”

Last year, while working on visualizations of a quantum-centric supercomputer with Jay Gambetta, vice president and lead scientist of IBM Quantum at the Thomas J. Watson Research Center in Yorktown Heights, N.Y., De Torres was inspired to contact Spectrum’s creative director, Mark Montgomery, with an idea.

“I really loved this process because I got to bring together two of my favorite clients to create something really special.” —Carl De Torres

“I thought, ‘You know, I think IEEE Spectrum would love to see this work,’” De Torres told me. “So with Jay’s permission, I gave Mark a 30-second pitch. Mark liked it and ran it by the editors, and they said that it sounded very promising.” De Torres, members of the IBM Quantum team, and Spectrum editors had a call to brainstorm what the article could be. “From there everything quickly fell into place, and I worked with Spectrum and the IBM Quantum team on a visual approach to the story,” De Torres says.

As for the text, we knew it would take a deft editorial hand to help the authors explain what amounts to the peanut butter and chocolate of advanced computing. Fortunately for us, and for you, dear reader, Associate Editor Dina Genkina has a doctorate in atomic physics, in the subfield of quantum simulation. As Genkina explained to me, that speciality is “adjacent to quantum computing, but not quite the same—it’s more like the analog version of QC that’s not computationally complete.”

Genkina was thrilled to work with De Torres to make the technical illustrations both accurate and edifying. Spectrum prides itself on its tech illustrations, which De Torres notes are increasingly rare in the space-constrained era of mobile-media consumption.

“Working with Carl was so exciting,” Genkina says. “It was really his vision that made the article happen, and the scope of his ambition for the story was at times a bit terrifying. But it’s the kind of story where the illustrations make it come to life.”

De Torres was happy with the collaboration, too. “I really loved this process because I got to bring together two of my favorite clients to create something really special.”

This article appears in the September 2024 print issue.

Back in June 2022, Oak Ridge National Laboratory debuted Frontier—the world’s most powerful supercomputer. Frontier can perform a billion billion calculations per second. And yet there are computational problems that Frontier may never be able to solve in a reasonable amount of time.

Some of these problems are as simple as factoring a large number into primes. Others are among the most important facing Earth today, like quickly modeling complex molecules for drugs to treat emerging diseases, and developing more efficient materials for carbon capture or batteries.

However, in the next decade, we expect a new form of supercomputing to emerge unlike anything prior. Not only could it potentially tackle these problems, but we hope it’ll do so with a fraction of the cost, footprint, time, and energy. This new supercomputing paradigm will incorporate an entirely new computing architecture, one that mirrors the strange behavior of matter at the atomic level—quantum computing.

For decades, quantum computers have struggled to reach commercial viability. The quantum behaviors that power these computers are extremely sensitive to environmental noise, and difficult to scale to large enough machines to do useful calculations. But several key advances have been made in the last decade, with improvements in hardware as well as theoretical advances in how to handle noise. These advances have allowed quantum computers to finally reach a performance level where their classical counterparts are struggling to keep up, at least for some specific calculations.

For the first time, we here at IBM can see a path toward useful quantum computers, and we can begin imagining what the future of computing will look like. We don’t expect quantum computing to replace classical computing. Instead, quantum computers and classical computers will work together to run computations beyond what’s possible on either alone. Several supercomputer facilities around the world are already planning to incorporate quantum-computing hardware into their systems, including Germany’s Jupiter, Japan’s Fugaku, and Poland’s PSNC. While it has previously been called hybrid quantum-classical computing, and may go by other names, we call this vision quantum-centric supercomputing.

A Tale of Bits and Qubits

At the heart of our vision for a quantum-centric supercomputer is the quantum hardware, which we call a quantum processing unit (QPU). The power of the QPU to perform better than classical processing units in certain tasks comes from an operating principle that’s fundamentally different, one rooted in the physics of quantum mechanics.

In the standard or “classical” model of computation, we can reduce all information to strings of binary digits, bits for short, which can take on values of either 0 or 1. We can process that information using simple logic gates, like AND, OR, NOT, and NAND, which act on one or two bits at a time. The “state” of a classical computer is determined by the states of all its bits. So, if you have N bits, then the computer can be in just one of 2^N states.

But a quantum computer has access to a much richer repertoire of states during computation. A quantum computer also has bits. But instead of just 0 and 1, its quantum bits— qubits—via a quantum property known as superposition, represent 0, 1, or a linear combination of both. While a digital computer can be in just one of those 2^N states, a quantum computer can be in many logical states at once during the computation. And the superpositions the different qubits are in can be correlated with one another in a fundamental way, thanks to another quantum property known as entanglement. At the end of the computation, the qubit assumes just one state, chosen based on probabilities generated during the running of the quantum algorithm.

It’s not obvious how this computing paradigm can outperform the classical one. But in 1994, Peter Shor, a mathematician at MIT, discovered an algorithm that, using the quantum-computing paradigm, could divide large numbers into their prime factors exponentially faster than the best classical algorithm. Two years later, Lov Grover discovered a quantum algorithm that could find a particular entry in a database much faster than a classical one could.

Perhaps most importantly, since quantum computers follow the laws of quantum mechanics, they are the right tool for simulating the fundamentally quantum phenomena of our world, such as molecular interactions for drug discovery or materials design.

The Quantum-Centric Supercomputer’s Center

Before we can build a quantum-centric supercomputer, we have to make sure it’s capable of doing something useful. Building a capable enough QPU relies on constructing hardware that can re-create counterintuitive quantum behaviors.

Here at IBM, the basic building block of a quantum computation—the qubit—is made out of superconducting components. Each physical qubit consists of two superconducting plates, which act as a capacitor, wired to components called Josephson junctions, which act as a special lossless, nonlinear inductor.

The current flowing across Josephson junctions is quantized—fixed to discrete values. The Josephson junctions ensure that only two of those values (or their superpositions) are realistically accessible. The qubit is encoded in two current levels, one representing a 0, the other a 1. But, as mentioned, the qubit can also exist in a superposition of the 0 and 1 states.

Because superconductors need frigid temperatures to maintain superconductivity, the qubits and some of their control circuitry are held inside a specialty liquid-helium fridge called a dilution refrigerator.

We change the qubit states and couple qubits together with quantum instructions, commonly known as gates. These are a series of specially crafted microwave waveforms. A QPU includes all of the hardware responsible for accepting a set of quantum instructions—called a quantum circuit—and returning a single output represented by a binary string. The QPU includes the qubits plus components that amplify signals, the control electronics, and the classical computation required for tasks such as holding the instructions in memory, accumulating and separating signals from noise, and creating single binary outputs. We etch components like qubits, resonators for readouts, output filters, and quantum buses into a superconducting layer deposited on top of a silicon chip.

But it’s a challenge trying to control qubits at the supersensitive quantum level. External noise, noise from the electronics, and cross talk between control signals for different qubits all destroy the fragile quantum properties of the qubits. Controlling these noise sources has been key in reaching the point where we can envision useful quantum-centric supercomputers.

Getting the Quantum Stuff up to Snuff

No one has yet conclusively demonstrated quantum advantage—that is, a quantum computer that outperforms the best classical one on a real-world relevant task. Demonstrating true quantum advantage would herald a new era of computing, where previously intractable tasks would now be within reach.

Before we can approach this grandiose goal, we have to set our sights a bit lower, to a target we call quantum utility. Quantum utility is the ability of quantum hardware to outperform brute-force classical calculations of a quantum circuit. In other words, it’s the point where quantum hardware is better at doing quantum computations than a traditional computer is.

This may sound underwhelming, but it is a necessary stepping-stone on the way to quantum advantage. In recent years, the quantum community has finally reached this threshold. Demonstrating quantum utility of our QPU, which we did in 2023, has convinced us that our quantum hardware is advanced enough to merit being built into a quantum-centric supercomputer. Achieving this milestone has taken a combination of advances, including both hardware and algorithmic improvements.

Since 2019, we’ve been incorporating advances in semiconductor fabrication to introduce 3D integration to our chips. This gave us access to qubits from a controller chip placed below the qubit plane to reduce the wiring on the chip, a potential source of noise. We also introduced readout multiplexing, which allows us to access the information from several qubits with a single wire, drastically reducing the amount of hardware we have to put in the dilution refrigerator.

In 2023, we implemented a new way to perform quantum gates—the steps of a program that change the value of the qubits—on our hardware, using components called tunable couplers. Previously, we prevented cross talk by fabricating the qubits that respond to different frequencies so that they wouldn’t react to microwave pulses meant for other qubits. But this made it too difficult for the qubits to perform the essential task of talking to one another, and it also made the processors slow. With tunable couplers, we don’t need the frequency-specific fabrication. Instead, we introduced a sort of “on-off” switch, using magnetic fields to decide whether or not a qubit should talk to another qubit. The result: We virtually eliminated cross-talk errors between qubits, allowing us to run much faster, more reliable gates.

As our hardware improved, we also demonstrated that we could deal with some noise using an error mitigation algorithm. Error mitigation can be done in many ways. In our case, we run quantum programs, analyze how the noise in our system changes the program outputs, and then create a noise model. Then we can use classical computing and our noise model to recover what a noise-free result would look like. The surrounding hardware and software of our quantum computer therefore includes classical computing capable of performing error mitigation, suppression, and eventually, error correction.

Alongside ever-improving hardware advances, we teamed up with the University of California, Berkeley, to demonstrate in 2023 that a quantum computer running our 127-qubit quantum chip, Eagle, could run circuits beyond the ability of brute-force classical simulation—that is, methods where the classical computer exactly simulates the quantum computer in order to run the circuit, reaching quantum utility. And we did so for a real condensed-matter physics problem—namely, finding the value of a property called magnetization for a system of simplified atoms with a structure that looked like the layout of our processors’ qubits.

Left: A quantum processing unit is more than just a chip. It includes the interconnects, amplifiers, and signal filtering. It also requires the classical hardware, including the room-temperature classical computers needed to receive and apply instructions and return outputs. Right: At the heart of an IBM quantum computer is a multilayer semiconductor chip etched with superconducting circuits. These circuits comprise the qubits used to perform calculations. Chips are divided into a layer with the qubits, a layer with resonators for readout, and multiple layers of wiring for input and output.

Error Correction to the Rescue

We were able to demonstrate the ability of our quantum hardware outperforming brute-force classical simulation without leveraging the most powerful area of quantum-computing theory: quantum error correction.

Unlike error mitigation, which deals with noise after a computation, quantum error correction can remove noise as it arises during the process. And it works for a more general kind of noise; you don’t need to figure out a specific noise model first. Plus, while error mitigation is limited in its ability to scale as the complexity of quantum circuits grows, error correction will continue to work at large scales.

Error Correction

But quantum error correction comes at a huge cost: It requires more qubits, more connectivity, and more gates. For every qubit you want to compute with, you may need many more to enable error correction. Recent advances in improving hardware and finding better error-correcting codes have allowed us to envision an error-corrected supercomputer that can make those costs worthwhile.

Quantum error-correcting schemes are a bit more involved than error correction in traditional binary computers. To work at all, these quantum schemes require that the hardware error rate is below a certain threshold. Since quantum error correction’s inception, theorists have devised new codes with more relaxed thresholds, while quantum-computer engineers have developed better-performing systems. But there hasn’t yet been a quantum computer capable of using error correction to perform large-scale calculations.

Meanwhile, error-correction theory has continued to advance. One promising finding by Moscow State University physicists Pavel Panteleev and Gleb Kalachev inspired us to pursue a new kind of error-correcting code for our systems. Their 2021 paper demonstrated the theoretical existence of “good codes,” codes where the number of extra qubits required to perform error correction scales more favorably.

This led to an explosion of research into a family of codes called quantum low-density parity check codes, or qLDPC codes. Earlier this year, our team published a qLDPC code with an error threshold high enough that we could conceivably implement it on near-term quantum computers; the amount of required connectivity between qubits was only slightly beyond what our hardware already supplies. This code would need only a tenth the number of qubits as previous methods to achieve error correction at the same level.

These theoretical developments allow us to envision an error-corrected quantum computer at experimentally accessible scales, provided we can connect enough quantum processing power together, and leverage classical computing as much as possible.

Hybrid Classical-Quantum Computers for the Win

To take advantage of error correction, and to reach large enough scales to solve human-relevant problems with quantum computers, we need to build larger QPUs or connect multiple QPUs together. We also need to incorporate classical computing with the quantum system.

Quantum-centric supercomputers will include thousands of error-corrected qubits to unlock the full power of quantum computers. Here’s how we’ll get there.

2024

Heron

→ 156 qubits

→ 5K gates before errors set in

2025

Flamingo

→ Introduce l-couplers between chips

→ Connect 7 chips for 7 x 156 = 1,092 qubits

→ 5K gates before errors set in

2027

Flamingo

→ l-couplers between chips

→ 7 x 156 = 1,092 qubits

→ Improved hardware and error mitigation

→ 10K gates before errors set in

2029

Starling

→ 200 qubits

→ l-, m-, and c-couplers combined

→ Error correction

→ 100M gates

2030

BlueJay

→ 2,000 qubits

→ Error correction

→ 1B gates

Last year, we released a machine we call the IBM Quantum System Two, which we can use to start prototyping error mitigation and error correction in a scalable quantum computing system. System Two relies on larger, modular cryostats, allowing us to place multiple quantum processors into a single refrigerator with short-range interconnects, and then combine multiple fridges into a bigger system, kind of like adding more racks to a traditional supercomputer.

Along with the System Two release, we also detailed a 10-year plan for realizing our vision. Much of the early hardware work on that road map has to do with interconnects. We’re still developing the interconnects required to connect quantum chips into larger chips like Lego blocks, which we call m-couplers. We’re also developing interconnects to transfer quantum information between more distant chips, called l-couplers. We hope to prototype both m- and l-couplers by the end of this year. We’re also developing on-chip couplers that link qubits on the same chip that are more distant than their nearest neighbors—a requirement of our newly developed error-correction code. We plan to deliver this c-coupler by the end of 2026. In the meantime, we’ll be improving error mitigation so that by 2028, we can run a quantum program across seven parallel quantum chips, each chip capable of performing up to 15,000 accurate gates before the errors set in, on 156 qubits.

We’re also continuing to advance error correction. Our theorists are always looking for codes that require fewer extra qubits for more error-correcting power and allow for higher error thresholds. We must also determine the best way to run operations on information that’s encoded into the error-correcting code, and then decode that information in real time. We hope to demonstrate those by the end of 2028. That way, in 2029, we can debut our first quantum computer incorporating both error mitigation and error correction that can run up to 100 million gates until the errors take hold, on 200 qubits. Further advances in error correction will allow us to run a billion gates on 2,000 qubits by 2033.

Knitting Together a Quantum-Centric Supercomputer

The ability to mitigate and correct errors removes a major roadblock in the way of full-scale quantum computing. But we still don’t think it’ll be enough to tackle the largest, most valuable problems. For that reason, we’ve also introduced a new way of running algorithms, where multiple quantum circuits and distributed classical computing are woven together into a quantum-centric supercomputer.

Many envision the “quantum computer” as a single QPU, working on its own to run programs with billions of operations on millions of physical qubits. Instead, we envision computers incorporating multiple QPUs, running quantum circuits in parallel with distributed classical computers.

Combining the strengths of quantum and classical

Quantum-centric supercomputing leverages quantum and classical resources in parallelized workloads to run computations larger than what was possible before. A quantum-centric supercomputer is a system optimized to orchestrate work across the quantum computers and advanced classical compute clusters in the same data center.

Recent work has demonstrated techniques that let us run quantum circuits much more efficiently by incorporating classical computing with quantum processing. These techniques, called circuit knitting, break down a single quantum-computing problem into multiple quantum-computing problems and then run them in parallel on quantum processors. And then a combination of quantum and classical computers knit the circuit results together for the final answer.

Another technique uses the classical computer to run all but the core, intrinsically quantum part of the calculation. It is this last vision that we believe will realize quantum advantage first.

Therefore, a quantum computer doesn’t just include one quantum processor, its control electronics, and its dilution refrigerator—it also includes the classical processing required to perform error correction, and error mitigation.

We haven’t realized a fully integrated quantum-centric supercomputer yet. But we’re laying the groundwork with System Two, and Qiskit, our full-stack quantum-computing software for running large quantum workloads. We are building middleware capable of managing circuit knitting, and of provisioning the appropriate computing resources when and where they’re required. The next step is to mature our hardware and software infrastructure so that quantum and classical can extend one another to do things beyond the capabilities of either.

Today’s quantum computers are now scientific tools capable of running programs beyond the brute-force ability of classical simulation, at least when simulating certain quantum systems. But we must continue improving both our quantum and classical infrastructure so that, combined, it’s capable of speeding up solutions for problems relevant to humanity. With that in mind, we hope that the broader computing community will continue researching new algorithms incorporating circuit knitting, parallelized quantum circuits, and error mitigation in order to find use cases that can benefit from quantum in the near term.

And we look forward to a day when the Top 500 list of most powerful supercomputers will include machines that have quantum processors at their hearts.

Today, almost all data on the Internet, including bank transactions, medical records, and secure chats, is protected with an encryption scheme called RSA (named after its creators Rivest, Shamir, and Adleman). This scheme is based on a simple fact—it is virtually impossible to calculate the prime factors of a large number in a reasonable amount of time, even on the world’s most powerful supercomputer. Unfortunately, large quantum computers, if and when they are built, would find this task a breeze, thus undermining the security of the entire Internet.

Luckily, quantum computers are only better than classical ones at a select class of problems, and there are plenty of encryption schemes where quantum computers don’t offer any advantage. Today, the U.S. National Institute of Standards and Technology (NIST) announced the standardization of three post-quantum cryptography encryption schemes. With these standards in hand, NIST is encouraging computer system administrators to begin transitioning to post-quantum security as soon as possible.

“Now our task is to replace the protocol in every device, which is not an easy task.” —Lily Chen, NIST

These standards are likely to be a big element of the Internet’s future. NIST’s previous cryptography standards, developed in the 1970s, are used in almost all devices, including Internet routers, phones, and laptops, says Lily Chen, head of the cryptography group at NIST who lead the standardization process. But adoption will not happen overnight.

“Today, public key cryptography is used everywhere in every device,” Chen says. “Now our task is to replace the protocol in every device, which is not an easy task.”

Why we need post-quantum cryptography now

Most experts believe large-scale quantum computers won’t be built for at least another decade. So why is NIST worried about this now? There are two main reasons.

First, many devices that use RSA security, like cars and some IoT devices, are expected to remain in use for at least a decade. So they need to be equipped with quantum-safe cryptography before they are released into the field.

“For us, it’s not an option to just wait and see what happens. We want to be ready and implement solutions as soon as possible.” —Richard Marty, LGT Financial Services

Second, a nefarious individual could potentially download and store encrypted data today, and decrypt it once a large enough quantum computer comes online. This concept is called “harvest now, decrypt later“ and by its nature, it poses a threat to sensitive data now, even if that data can only be cracked in the future.

Security experts in various industries are starting to take the threat of quantum computers seriously, says Joost Renes, principal security architect and cryptographer at NXP Semiconductors. “Back in 2017, 2018, people would ask ‘What’s a quantum computer?’” Renes says. “Now, they’re asking ‘When will the PQC standards come out and which one should we implement?’”

Richard Marty, chief technology officer at LGT Financial Services, agrees. “For us, it’s not an option to just wait and see what happens. We want to be ready and implement solutions as soon as possible, to avoid harvest now and decrypt later.”

NIST’s competition for the best quantum-safe algorithm

NIST announced a public competition for the best PQC algorithm back in 2016. They received a whopping 82 submissions from teams in 25 different countries. Since then, NIST has gone through 4 elimination rounds, finally whittling the pool down to four algorithms in 2022.

This lengthy process was a community-wide effort, with NIST taking input from the cryptographic research community, industry, and government stakeholders. “Industry has provided very valuable feedback,” says NIST’s Chen.

These four winning algorithms had intense-sounding names: CRYSTALS-Kyber, CRYSTALS-Dilithium, Sphincs+, and FALCON. Sadly, the names did not survive standardization: The algorithms are now known as Federal Information Processing Standard (FIPS) 203 through 206. FIPS 203, 204, and 205 are the focus of today’s announcement from NIST. FIPS 206, the algorithm previously known as FALCON, is expected to be standardized in late 2024.

The algorithms fall into two categories: general encryption, used to protect information transferred via a public network, and digital signature, used to authenticate individuals. Digital signatures are essential for preventing malware attacks, says Chen.

Every cryptography protocol is based on a math problem that’s hard to solve but easy to check once you have the correct answer. For RSA, it’s factoring large numbers into two primes—it’s hard to figure out what those two primes are (for a classical computer), but once you have one it’s straightforward to divide and get the other.

“We have a few instances of [PQC], but for a full transition, I couldn’t give you a number, but there’s a lot to do.” —Richard Marty, LGT Financial Services

Two out of the three schemes already standardized by NIST, FIPS 203 and FIPS 204 (as well as the upcoming FIPS 206), are based on another hard problem, called lattice cryptography. Lattice cryptography rests on the tricky problem of finding the lowest common multiple among a set of numbers. Usually, this is implemented in many dimensions, or on a lattice, where the least common multiple is a vector.

The third standardized scheme, FIPS 205, is based on hash functions—in other words, converting a message to an encrypted string that’s difficult to reverse

The standards include the encryption algorithms’ computer code, instructions for how to implement it, and intended uses. There are three levels of security for each protocol, designed to future-proof the standards in case some weaknesses or vulnerabilities are found in the algorithms.

Lattice cryptography survives alarms over vulnerabilities

Earlier this year, a pre-print published to the arXiv alarmed the PQC community. The paper, authored by Yilei Chen of Tsinghua University in Beijing, claimed to show that lattice-based cryptography, the basis of two out of the three NIST protocols, was not, in fact, immune to quantum attacks. On further inspection, Yilei Chen’s argument turned out to have a flaw—and lattice cryptography is still believed to be secure against quantum attacks.

On the one hand, this incident highlights the central problem at the heart of all cryptography schemes: There is no proof that any of the math problems the schemes are based on are actually “hard.” The only proof, even for the standard RSA algorithms, is that people have been trying to break the encryption for a long time, and have all failed. Since post-quantum cryptography standards, including lattice cryptography, are newer, there is less certainty that no one will find a way to break them.

That said, the failure of this latest attempt only builds on the algorithm’s credibility. The flaw in the paper’s argument was discovered within a week, signaling that there is an active community of experts working on this problem. “The result of that paper is not valid, that means the pedigree of the lattice-based cryptography is still secure,” says NIST’s Lily Chen (no relation to Tsinghua University’s Yilei Chen). “People have tried hard to break this algorithm. A lot of people are trying, they try very hard, and this actually gives us confidence.”

NIST’s announcement is exciting, but the work of transitioning all devices to the new standards has only just begun. It is going to take time, and money, to fully protect the world from the threat of future quantum computers.

“We’ve spent 18 months on the transition and spent about half a million dollars on it,” says Marty of LGT Financial Services. “We have a few instances of [PQC], but for a full transition, I couldn’t give you a number, but there’s a lot to do.”

This is a sponsored article brought to you by BESydney.

Australia plays a crucial role in global scientific endeavours, with a significant contribution recognized and valued worldwide. Despite comprising only 0.3 percent of the world’s population, it has contributed over 4 percent of the world’s published research.

Renowned for collaboration, Australian scientists work across disciplines and with international counterparts to achieve impactful outcomes. Notably excelling in medical sciences, engineering, and biological sciences, Australia also has globally recognized expertise in astronomy, physics and computer science.

As the country’s innovation hub and leveraging its robust scientific infrastructure, world-class universities and vibrant ecosystem, Sydney is making its mark on this burgeoning industry.

The city’s commitment to quantum research and development is evidenced by its groundbreaking advancements and substantial government support, positioning it at the forefront of the quantum revolution.

Sydney’s blend of academic excellence, industry collaboration and strategic government initiatives is creating a fertile ground for cutting-edge quantum advancements.

Sydney’s quantum ecosystem

Sydney’s quantum industry is bolstered by the Sydney Quantum Academy (SQA), a collaboration between four top-tier universities: University of NSW Sydney (UNSW Sydney), the University of Sydney (USYD), University of Technology Sydney (UTS), and Macquarie University. SQA integrates over 100 experts, fostering a dynamic quantum research and development environment.

With strong government backing Sydney is poised for significant growth in quantum technology, with a projected A$2.2 billion industry value and 8,700 jobs by 2030. The SQA’s mission is to cultivate a quantum-literate workforce, support industry partnerships and accelerate the development of quantum technology.

Professor Hugh Durrant-Whyte, NSW Chief Scientist and Engineer, emphasizes Sydney’s unique position: “We’ve invested in quantum for 20 years, and we have some of the best people at the Quantum Academy in Sydney. This investment and talent pool make Sydney an ideal place for pioneering quantum research and attracting global talent.”

Key institutions and innovations

UNSW’s Centre of Excellence for Quantum Computation and Communication Technology is at the heart of Sydney’s quantum advancements. Led by Scientia Professor Michelle Simmons AO, the founder and CEO of Silicon Quantum Computing, this centre is pioneering efforts to develop the world’s first practical supercomputer. This team is at the vanguard of precision atomic electronics, pioneering the fabrication of devices in silicon that are pivotal for both conventional and quantum computing applications and they have created the narrowest conducting wires and the smallest precision transistors.

“We can now not only put atoms in place but can connect complete circuitry with atomic precision.” —Michelle Simmons, Silicon Quantum Computing

Simmons was named 2018 Australian of the Year and won the 2023 Prime Minister’s Prize for Science for her work in creating the new field of atomic electronics. She is an Australian Research Council Laureate Fellow, a Fellow of the Royal Society of London, the American Academy of Arts and Science, the American Association of the Advancement of Science, the UK Institute of Physics, the Australian Academy of Technology and Engineering and the Australian Academy of Science.

In response to her 2023 accolade, Simmons said: “Twenty years ago, the ability to manipulate individual atoms and put them where we want in a device architecture was unimaginable. We can now not only put atoms in place but can connect complete circuitry with atomic precision—a capability that was developed entirely in Australia.”

The Design Futures Lab at UNSW in Sydney, Australia, is a hands-on teaching and research lab that aims to inspire exploration, innovation, and research into fabrication, emerging technologies, and design theories.UNSW

Government and industry support

In April 2024, the Australian Centre for Quantum Growth program, part of the National Quantum Strategy, provided a substantial four-year grant to support the quantum industry’s expansion in Australia. Managed by the University of Sydney, the initiative aims to establish a central hub that fosters industry growth, collaboration, and research coordination.

This centre will serve as a primary resource for the quantum sector, enhancing Australia’s global competitiveness by promoting industry-led solutions and advancing technology adoption both domestically and internationally. Additionally, the centre will emphasise ethical practices and security in the development and application of quantum technologies.

Additionally, Sydney hosts several leading quantum startups, such as Silicon Quantum Computing, Quantum Brilliance, Diraq and Q-CTRL, which focus on improving the performance and stability of quantum systems.

Educational excellence

Sydney’s universities are globally recognized for their contributions to quantum research. They nurture future quantum leaders, and their academic prowess attracts top talent and fosters a culture of innovation and collaboration.

Sydney hosts several leading quantum startups, such as Silicon Quantum Computing, Quantum Brilliance, Diraq, and Q-CTRL, which focus on improving the performance and stability of quantum systems.

The UNSW Sydney is, one of Sydney’s universities, ranked among the world’s top 20 universities, and boasts the largest concentration of academics working in AI and quantum technologies in Australia.

UNSW Sydney Professor Toby Walsh is Laureate Fellow and Scientia Professor of Artificial Intelligence at the Department of Computer Science and Engineering at the University of New South Wales. He explains the significance of this academic strength: “Our students and researchers are at the cutting edge of quantum science. The collaborative efforts within Sydney’s academic institutions are creating a powerhouse of innovation that is driving the global quantum agenda.”

Sydney’s strategic investments and collaborative efforts in quantum technology have propelled the city to the forefront of this transformative field. With its unique and vibrant ecosystem, a blend of world-leading institutions, globally respected talent and strong government and industry support, Sydney is well-positioned to lead the global quantum revolution for the benefit of all. For more information on Sydney’s science and engineering industries visit besydney.com.au.

Quantum computing is a devilishly complex technology, with many technical hurdles impacting its development. Of these challenges two critical issues stand out: miniaturization and qubit quality.

IBM has adopted the superconducting qubit road map of reaching a 1,121-qubit processor by 2023, leading to the expectation that 1,000 qubits with today’s qubit form factor is feasible. However, current approaches will require very large chips (50 millimeters on a side, or larger) at the scale of small wafers, or the use of chiplets on multichip modules. While this approach will work, the aim is to attain a better path toward scalability.

Now researchers at MIT have been able to both reduce the size of the qubits and done so in a way that reduces the interference that occurs between neighboring qubits. The MIT researchers have increased the number of superconducting qubits that can be added onto a device by a factor of 100.

“We are addressing both qubit miniaturization and quality,” said William Oliver, the director for the Center for Quantum Engineering at MIT. “Unlike conventional transistor scaling, where only the number really matters, for qubits, large numbers are not sufficient, they must also be high-performance. Sacrificing performance for qubit number is not a useful trade in quantum computing. They must go hand in hand.”

The key to this big increase in qubit density and reduction of interference comes down to the use of two-dimensional materials, in particular the 2D insulator hexagonal boron nitride (hBN). The MIT researchers demonstrated that a few atomic monolayers of hBN can be stacked to form the insulator in the capacitors of a superconducting qubit.

Just like other capacitors, the capacitors in these superconducting circuits take the form of a sandwich in which an insulator material is sandwiched between two metal plates. The big difference for these capacitors is that the superconducting circuits can operate only at extremely low temperatures—less than 0.02 degrees above absolute zero (-273.15 °C).

Superconducting qubits are measured at temperatures as low as 20 millikelvin in a dilution refrigerator.Nathan Fiske/MIT

In that environment, insulating materials that are available for the job, such as PE-CVD silicon oxide or silicon nitride, have quite a few defects that are too lossy for quantum computing applications. To get around these material shortcomings, most superconducting circuits use what are called coplanar capacitors. In these capacitors, the plates are positioned laterally to one another, rather than on top of one another.

As a result, the intrinsic silicon substrate below the plates and to a smaller degree the vacuum above the plates serve as the capacitor dielectric. Intrinsic silicon is chemically pure and therefore has few defects, and the large size dilutes the electric field at the plate interfaces, all of which leads to a low-loss capacitor. The lateral size of each plate in this open-face design ends up being quite large (typically 100 by 100 micrometers) in order to achieve the required capacitance.

In an effort to move away from the large lateral configuration, the MIT researchers embarked on a search for an insulator that has very few defects and is compatible with superconducting capacitor plates.

“We chose to study hBN because it is the most widely used insulator in 2D material research due to its cleanliness and chemical inertness,” said colead author Joel Wang, a research scientist in the Engineering Quantum Systems group of the MIT Research Laboratory for Electronics.

On either side of the hBN, the MIT researchers used the 2D superconducting material, niobium diselenide. One of the trickiest aspects of fabricating the capacitors was working with the niobium diselenide, which oxidizes in seconds when exposed to air, according to Wang. This necessitates that the assembly of the capacitor occur in a glove box filled with argon gas.

While this would seemingly complicate the scaling up of the production of these capacitors, Wang doesn’t regard this as a limiting factor.

“What determines the quality factor of the capacitor are the two interfaces between the two materials,” said Wang. “Once the sandwich is made, the two interfaces are “sealed” and we don’t see any noticeable degradation over time when exposed to the atmosphere.”

This lack of degradation is because around 90 percent of the electric field is contained within the sandwich structure, so the oxidation of the outer surface of the niobium diselenide does not play a significant role anymore. This ultimately makes the capacitor footprint much smaller, and it accounts for the reduction in cross talk between the neighboring qubits.

“The main challenge for scaling up the fabrication will be the wafer-scale growth of hBN and 2D superconductors like [niobium diselenide], and how one can do wafer-scale stacking of these films,” added Wang.

Wang believes that this research has shown 2D hBN to be a good insulator candidate for superconducting qubits. He says that the groundwork the MIT team has done will serve as a road map for using other hybrid 2D materials to build superconducting circuits.