From binary computing to quantum AI

The first computer using binary logic was invented in Europe. The Z1, designed and constructed by Konrad Zuse in Berlin in the late 1930’s, laid the foundation for high-performance computing for a century to come. The most powerful computers of today still use this rather primitive logic of on or off, of zeros becoming ones and vice versa. Now, this simple human-invented logic is being augmented by a more efficient approach based on the primordial force of the universe – quantum computing.

The quantum mechanical world is unruly, odd, and it certainly goes against everyday experience. Here, nothing is definite – everything is based on probabilities and a good measure of fuzziness. The ”operating system of the universe” provides a major advantage for computing by transcending the binary limitations of classical computing by lifting the restriction of either zero or one. A quantum computer follows quantum logic, where a mixture of ones and zeros forms the basic information unit, enabling a leap in computational possibilities.

This allows for significantly more powerful solutions to certain types of problems – many of which are critical for advancing our society. Quantum computing has already moved from laboratory to real-world settings. The inherent power of this new computing paradigm has been demonstrated as proofs-of-concept in such a convincing manner, that Draghi called it ”the next trailblazing innovation in the computing field” [1].

Alongside quantum computing, artificial intelligence (AI) is presently the computing paradigm. I find it invigorating that after a long period of steady incremental progress on the conventional computing side, we now have two new powerful approaches to simulation and computational RDI. Interestingly, AI and quantum computing synergistically enhance each other, creating a powerful feedback loop that can accelerate advancements in both fields.

Quantum AI is a rapidly developing field, motivated by its potential to recognise patterns both faster and more accurately than purely classical AI methods. In a synergistic two-way, AI methods for enhancing quantum computing are becoming crucial tools for extracting maximum utility, shortening the waiting time to true quantum advantage.

Quantum AI is, like all quantum computing applications, still at the exploratory stage. For exploration to be possible, it is essential for end-users to have access to cutting-edge quantum hardware. Recognising the importance of quantum computing at an early stage, the EuroHPC Joint Undertaking has recently procured eight different European quantum computers to be integrated with high-performance computing (HPC) systems around the continent – in Germany, France, Poland, Czechia, Italy, and Spain with two more systems coming to Luxembourg and the Netherlands.

From a quantum AI perspective, VLQ, the quantum computer of the LUMI-Q consortium located in Ostrava, Czechia, is especially exciting. It is based on a novel “star-shape” layout of the quantum processing unit, the QPU [2]. This design is ideal for creating so-called entangled states. The weirdness of entanglement led Einstein to call it “spooky action at a distance” – in the case of quantum AI, it forms the basis for outperforming purely classical methods.

The happy marriage of quantum and AI is further ensured by the EuroHPC JU AI Factories initiative [3]. Alongside VLQ and other EuroHPC platforms such as EuroQCS-Spain, EuroQCS-Italy and MeluXina-Q, the experimental platform of the LUMI AI Factory [4], LUMI-IQ, will provide a world-leading HPC+AI+QC platform for European end-users in both academia and industry. Through EuroHPC, the entire European RDI community can thus soon get their hands dirty, digging into the promise of quantum-enhanced AI.

As we celebrate the United Nations International Year of Quantum Science and Technology, it is time to take quantum computing to the next level of utility. Looking to the future, Europe has a genuine opportunity to be world-leading on both the quantum for AI and AI for quantum fronts. This does require long-term support for bringing the European hybrid AI+QC computing infrastructure up to international speed, both on hardware and software sides.

We do have the means and talent, now it is just a matter of eking out the collective will and commencing the action to position Europe at the forefront of the quantum AI revolution!

Author: Mikael Johansson
Manager of Quantum Technologies
CSC – IT Center for Science Ltd, Finland

Image (c): IT4Innovations

References

[1] M. Draghi, ”The Future of European competitiveness Part B – In-depth analysis and recommendations.” (2024). Available at https://commission.europa.eu/topics/strengthening-european-competitiveness/eu-competitiveness-looking-ahead_en

[2] VLQ Quantum Computer. https://www.it4i.cz/en/infrastructure/vlq-quantum-computer

[3] ” Selection of the First Seven AI Factories to Drive Europe’s Leadership in AI” (2024). https://eurohpc-ju.europa.eu/selection-first-seven-ai-factories-drive-europes-leadership-ai-2024-12-10_en

[4] LUMI AI Factory. https://www.lumi-supercomputer.eu/lumi-ai-factory/