Quantum Technologies and Artificial Intelligence: Shaking the Very Foundations of Science

Since the 17^th century (perhaps earlier), the scientific method has been the foundation on which we humans have built our understanding of the world around us. More akin to a loose set of guiding principles than a rigorous method, it broadly involves observation of some phenomenon, an idea formalized into a hypothesis for why the phenomenon occurs, and some test(s) to prove whether the hypothesis is correct or not – though these steps may not occur in that order, there may be other steps or some might be skipped entirely.

Scientific Method Stumbles

During their Focus Talk (1 July 2026, 11:00 CEST) at the 75^th Lindau Nobel Laureate Meeting, a trio of Physics Nobel Laureates – David J. Gross (2004), Serge Haroche (2012) and Arthur B. McDonald (2015) – will highlight just how far this rather loose and distinctly human scientific method has taken us in our understanding of the universe. From the most infinitesimally small constituents of matter to the structure and evolution of the entire universe, they will argue that the scientific method has broadened humanity’s horizons across an extraordinary range of scales.

Yet there are fundamental questions about the nature of the universe and reality that remain stubbornly unanswered. One of the most irksome for physicists, and likely to be highlighted in the Focus Talk, is the schism between 1921 Nobel Prize in Physics recipient Albert Einstein’s general relativity (which perfectly models the behaviour of matter at large scales) and quantum mechanics (which does the same at the smallest of scales); the latter of which emerged through the efforts of a host of physicists, including Einstein himself, as well as other Physics Nobel recipients Max Planck (1918), Werner Heisenberg (1932), Erwin Schrödinger (1933) and many more.

Another one of these quantum trailblazers, Paul Dirac (1933 Nobel Prize in Physics), successfully merged quantum mechanics with special relativity (how space, time, mass and energy are related) in 1930, creating the Dirac equation. But uniting quantum mechanics with general relativity (which adds gravity to the picture) proved more challenging. In fact, it is still an open problem. Years have turned to decades have turned to almost a century trying to crack this puzzle. Even worse, meaningful progress towards finding a solution appears to have stalled. It could even be argued, and has been, that no theoretical progress has been made at all since the mid-1970s when the Standard Model of Particle Physics was completed.

AI to the Rescue?

To get the wheels of progress turning once again on this and other seemingly intractable problems – not just in physics but across the sciences – might new technology be the answer?

Already playing its part across a host of scientific areas is AI. Neural networks, for example, are a type of AI being wielded to extract information from rapidly flowing data streams generated at Big Science facilities like the Large Hadron Collider and James Webb Space Telescope. Another area they and other types of AI, including large language models (LLMs), are providing huge benefits already is health, where they are being employed to assist in disease detection and prediction, drug discovery and clinical decision making. And, more generally, LLMs are helping all types of researchers in desk-based research, academic writing and coding.

In a Panel Discussion (1 July, 15:30 CEST), Nobel Laureates Geoffrey Hinton (2024 Physics), John M. Jumper (2024 Chemistry), Anne L’Huillier (2023 Physics) and William D. Phillips (1997 Physics) will touch upon many of these applications. However, the conversation will no doubt go well beyond current uses to how it might effect the scientific process itself.

As two pioneers of the field, Hinton and Jumper’s opinions, in particular, will be fascinating. Since resigning from Google in 2023, Hinton has been a prominent critic of unregulated AI progress, warning that artificial general intelligence could catalyse mass unemployment and even represent an existential risk to humanity. In contrast, Jumper has been on record saying that he has been impressed with how responsibly scientists have used AlphaFold, a neural network he co-developed at Google DeepMind with Demis Hassabis for protein structure prediction, and how AI should be used to make science go faster and enable new discoveries.

But for young researchers, perhaps the most important area of discussion will be around what AI-driven science means to them. For example, if AI is as good at problem solving as a gifted postgraduate – as has certainly been shown in mathematics – but can produce results in the blink of an eye and cost a tiny fraction of a postgraduate’s salary, what role can they fulfil? Could it be that PhD and postgraduate jobs dry up or, as Hassabis recently argued, might it allow a single PhD student to produce the amount of work it would take a whole laboratory to produce today. A better idea of what the young researchers’ role should be in the age of AI will likely be welcomed by this year’s Young Scientists.

The Future Quantum Revolution

In the midst of the AI revolution, it is easy to forget another revolution has long been promised, but has yet to fully materialize: the quantum computing revolution. Quantum computers, so say optimists, will be so far beyond the capabilities of regular supercomputers that they will solve a vast swathe of problems, bringing transformative change to almost every sector: from optimizing the fertilizers that sustain our food supply to discovering drugs to cure all disease, or from designing better batteries for electric cars, to finally harnessing fusion power to provide the world with unlimited clean electricity.

On 2 July 2026 at 09:00 CEST, Nobel Laureates Michel H. Devoret (2025 Physics), John M. Martinis (2025 Physics) and Phillips once again will have a Panel Discussion on which of these and more potential world-changing discoveries are realistically attainable with future quantum computers.

As Devoret and Martinis have both led Google’s Quantum AI team at different times, they will offer an insider’s perspective of where the technology is and is likely heading, including the technical challenges of building useful and powerful quantum computers. Meanwhile, as an expert on quantum physics but also a polymath with a wide range of scientific interests, Phillips can weigh in on the implications for science and society of the quantum computing revolution – of which there are many.

Not least of these is the threat to cybersecurity. With a mature quantum computer having the potential to crack modern encryption protocols in minutes, all digital information would suddenly be exposed and at risk, from state secrets to bank account details. Even the most advanced quantum computers are nowhere near achieving the required order of millions of stable, error-corrected qubits needed to pose a threat. However, society is living on borrowed time. There is therefore a huge drive to develop and rapidly apply post-quantum cryptographic protocols, including the US National Institute of Standards and Technology (NIST)’s relatively recent release of its post-quantum cryptography standards, so that the world can plan ahead and build defences before it is too late.

Another significant threat relates to both quantum computers and AI: accessibility. The companies developing these technologies argue that they democratize science by offering open-access, cloud-based AI tools and quantum computing services. However, the hardware delivering them is concentrated in a clutch of powerful corporations, who also retain the most advanced versions of the technologies for themselves and aggressively recruit top talent from around the world to develop the technologies further. Whether this is healthy and sustainable for science and society remains an open question.

But perhaps the most fundamental questions both AI and quantum computing raise are deeply philosophical. If and when these black-box proprietary technologies hand us answers without us necessarily asking the questions, following how those technologies came to the answers or comprehending why those answers are correct, we will be in completely new territory. It will be fascinating to hear what panellists think words such as ‘idea’, ‘hypothesis’, ‘discovery’ and ‘understanding’ even mean in such a world, whether human understanding still has value, and if the scientific method still has legs.