Event

This conference was organised by Verdinum, and hosted at CY Tech in Pau, France.

I was kindly invited by Paul Gay, thanks again Paul!

Presentation

(From the website)

This conference brings together perspectives from computer engineers, environmental scientists, and sociologists on the links between artificial intelligence and its socio-environmental impacts. More specifically, it aims to take stock of frugal AI using existing measurement tools and implementation methods, while sociological concepts of technological non-neutrality enable us to anticipate long-term consequences.

Keynote Abstract

What if our software could work perfectly, even when the internet goes down? What if we could take back control of our data without sacrificing collaboration? In an era dominated by cloud giants and energy-intensive AI, Local-First software offers a compelling alternative: applications that prioritize the user’s device as the primary source of truth.

This keynote explores how Local-First principles, such as “Offline-First” operation, peer-to-peer synchronization, and conflict-free data types, can lead to more frugal, resilient, and ethical digital systems. By reducing dependence on centralized infrastructure, we not only reduce latency and energy consumption, but also give users greater privacy, better ownership of their data, and increased accessibility.