Series: Toward a Shared Language Between Humans and Machines

Large language models (LLMs) today produce fluent and coherent texts, to the point of giving the illusion of a real conversation. But behind this apparent mastery arises an interesting question: do machines really understand us, or are they only predicting words?

In a four-part series, we will explore one of the deepest challenges of artificial intelligence: building a true common ground between human meaning and machine logic. From cognitive limits to quantum horizons, and including the strategic role of humans, each article will shed light on one facet of these questions.

Part 1: Why Machines Still Struggle to Understand Us

We will begin with the fundamental fractures that separate human words from machine calculations. Whether it is the absence of lived experience, the absence of a “world,” or the gap between our human intentions and algorithmic operations, these fractures create a deep divide between our modes of communication.

Read the full article here.

Part 2: From Multimodality to World Models: Teaching Machines How to Experience

Can perception bridge this gap? We will explore multimodal systems, digital twins, and the world models that research is attempting to develop, all aiming to give machines a form of grounding in reality.

Read the full article here.

Part 3: Can Quantum Language Break the Limits of Simulation?

Beyond imitation, could quantum computing encode meaning into qubits and open a radically new path? This part will examine how the quantum processing of language combines the fascination of cutting-edge research with very real breakthroughs.

Read the full article here.

Part 4: Humans as Co-Creators: Humans and AI as Co-Creators: Toward a Shared Language

In the end, the future of human-machine communication does not lie in replacing humans with machines, but in co-creation. We will analyze the ethical, cultural, and strategic choices that will allow AI to be guided as a partner, not as a replacement.

Read the full article here.

Across these four essays, the idea is to draw out a central message: the challenge is not only technical. Over the course of the four articles, we will establish that it is also cultural, ethical, and political. Building a shared language with machines requires preserving what makes human intelligence unique, while daring to imagine new forms of collaboration.

I invite you to read this series with curiosity and an open mind. Each article is meant as both an exploration and an invitation to reflect, question, and discuss. I would be truly glad to hear your thoughts, reactions, or even disagreements on these topics. 

After all, building a shared language between humans and machines begins with dialogue between humans themselves.