Beyond the Hype: What I’m Looking for at Davos 2026

By Marianna Bonanome, Head of AI Data Partnerships & Adoption

Walking down the Davos Promenade in 2024 felt like witnessing a total transformation of the tech landscape. The chalets that once beamed with "Blockchain," "Metaverse," and "Web3" logos had been completely renovated. In their place stood "AI Houses," where even established tech giants were rebranding their core offerings as "AI-First".

The parallels to previous hype cycles are hard to miss. We’ve seen this movie before: the peak of inflated expectations, the FOMO, and the eventual settling into reality. As we prepare for Davos 2026, the real question isn't whether AI is a bubble, but rather what will remain once the hype cycle runs its course and these tools become part of the bedrock of global industry?

1. The Consolidation of the "Wrapper"

For the past couple of years, everyone claimed to have an "AI solution". This year, the ecosystem is undergoing an adjustment. We are seeing the rise of general-purpose foundation models that act more like operating systems and search engines than simple tools.

As major players integrate data cleaning, file analysis, and agentic workflows directly into their APIs, the niche "wrapper" startups—those providing a simple UI for someone else's model—are finding it harder to compete. We are moving toward a marketplace where models act as agents with direct API access to our most-used apps and data. The need for manual scripting to connect data to AI is slowly being replaced by models that act as the universal connector.

2. From Abstract Math to Practical Potential

This shift feels personal to me. Twenty years ago, during my doctoral research in quantum information science, I lived in a world of abstract mathematics. I was writing quantum algorithms to solve intractable mathematical problems, work that, at the time, felt far removed from any useful application.

This groundwork was a necessary foundation for the practical computational problem-solving we are finally beginning to explore today. We are no longer just theorizing about "what if"; we are exploring the intersection of AI’s pattern recognition and the quantum world. While the jury is still out on when or how "quantum advantage" might be fully realized, the potential of what some are calling "Physical Intelligence" is beginning to emerge. We’re still in the nascent, exploratory stages at the moment, but the possibility of pairing these technologies to tackle problems in the physical world is profound.

3. The Quantitative Frontier

While much of the public's attention is on language models, a quieter revolution is happening in quantitative AI. These models aren't predicting the next word like a chatbot; rather, they are being used to rapidly acquire deeper insights and process information across two critical domains: the physical and the mathematical.

By analyzing physical and chemical structures alongside impossibly complex mathematical calculations, these tools allow us to make data-driven predictions in highly technical industries, from materials science to cybersecurity. Some specific examples include:

• Navigation and Medical Diagnostics: We are seeing the development of navigation systems that use quantum sensors and AI to interpret the Earth’s magnetic field, providing a robust, resilient, alternative when GPS signals are jammed or spoofed. In medicine, advanced sensors and AI are being used in Magnetocardiography (MCG) to understand the heart’s magnetic activity, providing much faster and more accurate cardiac diagnosis than current diagnostic tools.
• Biopharma and Materials: In drug discovery, researchers use physics-informed AI to rapidly develop novel drug candidates and optimize entire drug portfolios, cutting years off the R&D timeline and saving millions of dollars. The same approach is reimagining the way researchers create new catalysts, alloys, battery chemistries, and materials to speed better, stronger, and safer products to market.
• Quantum-Secure Foundations: The Quantum Insider and others are framing 2026 as the "Year of Quantum Security" –t using quantitative AI to manage the complex mathematical architectures required for Post-Quantum Cryptography (PQC) and secure networking.

The real "winners" at Davos 2026 will be those who invest the time to guide these powerful tools with their existing expertise. For someone who started their career writing code for a non-existent quantum machine, watching the world catch up to the math is both a moment of pure, scientific delight and deep personal and professional satisfaction.

These powerful technologies are finally within our reach. The question is, what will we build with them and how will they benefit humanity? I hope to find the answer in Switzerland this week.