Quantum computing and AI – 
A superpower in the making?

By Carina Kiessling

This new Roland Berger publication explores the potential – and the problems – in bringing proven artificial intelligence (AI) together with emerging quantum technologies. Far from blue-sky thinking, many companies already have this massively lucrative prize firmly in their sights.

Highlighting the complementary strengths and weaknesses of the two technologies, detailed examples are provided of how AI is already changing the game and boosting efficiency in many areas. The discussion then turns to quantum computing and what it brings to the table. Developmental issues – and questions about the timeline – are tackled head-on. Questions are also asked about the purpose of and motivation for the export controls that have recently been imposed on quantum technologies in many countries. A panorama of the public and private quantum investment landscape is painted, highlighting the observable growth in private-sector commitments.

The publication concludes by discussing the possibilities that could arise from combining these two technologies, including concrete proposals for feasible hardware synergies as a point of departure.

Two into one?

As artificial intelligence continues to improve both business processes and our everyday lives, its capabilities are no longer called into question. The only question that do arise surround its sustainability and ramp-up curve, given the excessive energy requirements involved. Quantum computing, on the other hand, promises even greater breakthroughs but is still at the beginning of its journey. AI, in effect, is the “here and now” that could be dovetailed with the “still to come” that is quantum computing with practical and universal usage.

A question we are asked very often is: “Will quantum computing replace artificial intelligence, or could the two be combined to form a new superpower?”

To put it bluntly: The two technologies have very different strengths and therefore lend themselves to different use cases. AI excels at creativity and language and video processing, resulting in the current rise of GenAI and chatbots.

In contrast, what makes quantum computing so unique is that it can tackle complex problems that are beyond the capabilities of classical algorithms because they have an immense volume of possible outcomes. By harnessing the peculiar principles of quantum mechanics, such as superposition and entanglement, quantum computers can process information and solve problems faster than classical computers in areas such as cryptography, the search for new materials and highly complex optimization challenges.

That said, there are already promising concepts that are seeking to combine both technologies. In this article, we examine what might happen if AI and quantum computing were indeed combined and what impacts could be expected going forward.

AI comprises a series of well-established approaches, from machine learning to neural networks, deep neural networks and exploratory neuromorphic computing. Whenever people think of AI, the main applications are usually associated with the following six categories: natural language processing, speech recognition, computer vision, robotics and motion, AI planning and optimization, and knowledge treatment. Let us briefly examine all six in turn.

AI and its myriad applications already span the entire spectrum of industry markets. New use cases are evolving and being commercialized faster than ever. The section below outlines just a few tangible examples of AI applications that, once adopted, have given first movers the opportunity to realize significant value creation, business acceleration and competitive advantages.

Proven productivity gains with AI – The list is long!

Companies in various industries have recognized the disruptive impact of AI and are investing – or already using AI – in fields such as R&D, procurement and supply chain management. The following list showcases a selection of real-world AI applications:

• The R&D process: AI can be used to generate design solutions that adhere to specific goals and constraints. This approach permits the simultaneous exploration of multiple options, massive customization and design that is free from human biases. In addition, generative AI (GenAI) can generate, validate and manage requirement specifications while also tracking changes and dependencies that can greatly enhance the complexity of larger development modules in particular.
• Procurement: GenAI provides cost saving agents that identify category-specific cost saving levers and opportunities and translates them into concrete implementation measures. Furthermore, knowledge hubs significantly boost procurement efficiency by supporting and accelerating the research, generation and review of documents, for example.
• Supply chain management: Predictive analytics incorporating real-time data from various sources (such as market trends or customer feedback) enables companies to forecast demand (and fluctuations in demand) more accurately. In addition, AI facilitates advanced scenario planning and optimization, featuring multiple simulations based on different parameters and constraints.
• Manufacturing: By continuously analyzing equipment or product behavior, predictive maintenance can detect anomalies and identify failures before they occur. Furthermore, LLMs automate the creation and completion of PLC code in line with directives, best practices and predefined hardware constraints and requirements.
• Services: Chatbots already handle a large share of customer service and hotline tickets. By transcribing, categorizing and responding to customer inquiries, they shorten response times and reduce the number of person-hours needed. Moreover, LLMs can also automate the identification of sales potential and create and distribute personalized marketing material.

While AI's impact in areas such as productivity, innovation, services and enhanced decision-making is already reshaping whole industries, quantum technologies – such as quantum computing – are another promising technology with the potential to transform industries on a similar scale. Thinking back only a few years, AI was not yet widespread in any industry. Yet it has gained momentum with astonishing rapidity, transforming operations through advanced data analysis and automation. In similar fashion, quantum technologies too could take off very quickly, although the timeline and pace of its development remain uncertain.

Quantum computing is the largest aspect of the broader field of quantum technologies. (Other aspects include cryptography, communication, sensing and metrology.) As we have seen, quantum computing’s core strength lies in its ability to tackle highly complex issues with manifold outcomes and deliver very fast solutions. In contrast to AI, which relies mainly on powerful but traditional graphics processing units (GPUs), quantum computers harness advanced technologies – such as superconducting circuits, trapped ions and neutral atoms in highly isolated environments – to protect their fragile processing. While initial quantum algorithms have demonstrated promising results, the current challenge lies in scaling and improving the hardware to fully unlock and exploit the potential of quantum computing.

Several large technology companies and quantum start-ups already have development roadmaps in place for the years and, indeed, the decade ahead. As things stand, however, it remains unclear when quantum computers will reach sufficient quantum bit (qubit) numbers, low-enough error rates and adequate connectivity to take on high-impact problems. The current technology is still too error-prone to be able to crack encryption, for instance. As companies navigate and position themselves in the realm of quantum technologies, careful monitoring of all subfields is necessary, especially regarding the maturity of research and the projected time-to-market for the given quantum technology applications and solutions.

Export controls – Safety first, or a barrier to innovation?

Recent global export controls on quantum computers and related technologies, imposed by various countries, reflect a collective effort to regulate this emerging technology. The implication is that governments share concerns about its disruptive potential, not to mention its potential to be weaponized by militaries. The controls share similar specifications, such as limits on qubits and error rates, and seem to originate from undisclosed international discussions. However, the scientific rationale behind these regulations is not disclosed.

The uniformity of export controls seen in countries like the UK, France, Spain, the US and the Netherlands is the fruit of extended multilateral negotiations under what is known as the Wassenaar Arrangement. This framework, signed and backed by 42 member states, regulates the export of dual-use technologies that could have military applications. Yet despite claims that these limits are based on scientific analysis, the specific details and research supporting these decisions have not been made public.

Experts in the quantum industry have raised concerns about these measures, suggesting that they may be intended to limit the export of highly advanced quantum computers that are beyond the capabilities of simulation. However, it is important to note that the advanced nature of these quantum computers does not necessarily imply that they can indeed solve practical problems.

Moreover, it could be that such restrictions are what actually impedes innovation in this field. The readiness of quantum technology applications depends heavily on progress in research, and suffers from limitations imposed on the realization of quantum technology. Both factors will impact future roadmaps. It follows that external technical factors currently make it hard for first adopters in industry to plot reliable times to market for certain quantum applications.

Notwithstanding, the benefits that will undoubtedly result from breakthroughs in quantum technology development remain hugely attractive. Accordingly, many industrial players are more than willing to bet on quantum and invest in its usage. This being the case, it is important to take a closer look at the market potential and dynamics of quantum technologies.

Investment in quantum technologies and future prospects

Valued at around USD 3 billion in 2023, the global quantum market is expected to grow to USD 25 billion by 2028, implying a compound annual growth rate (CAGR) of 70% . Primary use cases driving the market for quantum technologies have so far been battery material development, navigation optimization, imaging and sensor technologies for autonomous driving, quantum-secured communication and the optimization of supply chains and customer relationship management, to name but a few.

Investments in quantum technologies, while previously dominated by public investments, have now also been picked up by private investors. Since 2021, deals worth around USD 2 billion per year have been linked to the latter group, tripling the total volume of private investments seen in 2020. Private investor interest (and easier access to quantum technology) is also underscored by the fact that 25 quantum technology firms are now publicly traded. Public spending nevertheless remains the primary source of capital: Worldwide, around USD 55 billion has, for example, been channeled into these technologies through government-backed funds such as the EU’s Horizon program, which committed around USD 7 billion in 2023.

Compared to the AI market, which is currently valued at USD 150 billion, quantum is still very much in its infancy. Yet this fact only underscores the opportunities that quantum holds out for both investors and companies. One sub-segment of AI is automotive AI, valued at around USD 3 billion in 2023 but expected to grow to USD 7.5 billion at a CAGR of 30% through 2029. Unlike quantum, private equity investments in automotive AI alone totaled USD 15 billion in 2023 – a number which again highlights how investors perceive the relative maturity of either technology. The volume of public investment spending likewise differs between the two technologies: For example, the EU’s Horizon program invested around USD 180 million across 28 AI and robotics projects in 2023, as the means for technological progress are mainly provided by the private sector.

Initial insights from efforts to combine AI and quantum computing point to benefits that range from enhanced efficiency to completely new fields of research. Despite the current challenges posed by error rates in quantum computers, there is significant interest in bringing the principles of quantum computing together with AI in areas from the mutual enhancement of both technologies to the execution of machine learning algorithms on quantum computers. Here are just three examples:

• Synergies between AI and quantum computing: Utilizing quantum-inspired machine learning algorithms could enhance AI techniques, and vice versa, using machine learning to optimize quantum error correction algorithms.
• Enhanced efficiency in image recognition: Initial results demonstrate that executing machine learning image recognition tasks on quantum computers can reduce the number of input parameters, leading to more compact processing.
• Quantum artificial intelligence: Since linguistics and cognition remain intractable to classical computer simulations, quantum natural language processing and quantum machine learning could be the new approach that paves the way to advances in this field.

Navigating the early stages of an uncertain but promising technological frontier

The combination of quantum computing and AI is in an early phase of active and exploratory research. As one would perhaps expect at such a stage, excitement and uncertainty are plentiful. And even as corporations and academic scientists dig deeper into hybrid quantum-AI approaches, the outcomes remain largely unclear, underscoring the experimental nature of this endeavor. Some experts are optimistic, expecting to see substantial advantages from the merging of these two technologies. Others remain cautious. And it is this dichotomy of views that reflects the current state of the art: a landscape of potential where the true benefits and applications of fusing quantum computing with AI are yet to be conclusively determined.

Synergies across AI and quantum hardware? The outlook is bright

Hardware synergies for the operation of both AI and quantum systems could be exploited across a range of technologies and components. As described above, attempts are being made to combine quantum technology and artificial intelligence to efficiently solve a wide range of problems. However, certain types of problems will still require separate treatment using AI-based methods, while others can doubtless be solved more efficiently on a quantum basis. This suggests that the two technological approaches may continue to coexist in the future. In this scenario, the question arises as to whether the hardware requirements for AI and quantum systems will be synergistic. Three technologies where potential for synergies appears to exist are proposed in the following list:

• Memory: Both quantum computing and artificial intelligence require large amounts of memory and storage. The preparation of quantum states requires vast swathes of digital storage, while AI systems need space for deep learning models and databases. In the case of quantum computing, memory readout time is also critical: When running quantum error-correction algorithms, bit-string sequences stored in memory must be applied with extreme precision on a timescale of a few hundred nanoseconds. Similarly, such fast readouts can also be very valuable when running real-time AI-enhanced pattern recognition algorithms.
• Network connections: Quantum computing requires high-speed connections to handle controlling signals from the sub-Kelvin superconducting qubit to the external microwave controllers, for example. At the same time, high-speed, low-latency network connections are key components in enabling AI systems to process real-time data quickly and with low error rates. Hence, both areas of technology require high-quality network components.
• Thermal management: Quantum computing and AI both require significant investments in cooling technology. Quantum computers must be kept at extremely low temperatures to operate effectively, while high-powered AI processors also generate large amounts of heat that must be extracted from the processors. Despite the 300 Kelvin gap in operating temperature regimes, both technologies require effective thermal anchoring, pre-cooling and passive cooling techniques – still more areas where synergies could be achieved in thermal control technology for both AI and quantum systems.

These three examples alone show that synergies are indeed possible between quantum and artificial intelligence hardware technology. And this realization opens up new opportunities off the beaten track for companies with portfolios that include either AI or quantum hardware. While caution is clearly in order, the window of opportunity that is now opening holds out immense possibilities. Having seen how AI alone is turning virtually every industry on its head, the potential gains from blending this game-changer with the barely conceivable capabilities of quantum computing are nothing short of mind-boggling. The conclusion? Far-reaching strategic decisions are looming in the very near future.

Roland Berger boasts experts who are highly experienced in advanced disciplines such as quantum technologies, photonics and optics, semiconductors, electrics and electronics, to name but a few. Better still, our experts also possess the skills needed to navigate clients through the new digital age, helping them to understand the applications of artificial intelligence, neural networks and the like. Their mandate is to support you in finding relevant ways to positively intersect technology business with industry demand and the latest research developments. Paired with our broad industry knowledge and an impressive track record of serving clients in all these fields, we would be happy to walk with you through a future of such thrilling possibilities. Feel free to reach out to any member of our Advanced Technology Team. We look forward to hearing from you!