Paul-Olivier Dehaye

Space-Proven AI

WINNER OF THIS YEAR: SGDL startup has received the Panoramai "Cosmos Award" and will be part of Panoramai Cohort 1 (mentoring and travel).

Revolutionary Insight: Physics-based reasoning systems proven in space operations offer a fundamentally new approach to AI security that matches the sophistication of autonomous multi-stage cyber attacks.

At Panoramai, Paul-Olivier Dehaye from SGDL revealed how space station robotics technology is revolutionizing AI cybersecurity. Using the analogy of astronauts « fishing » for incoming Dragon modules with robotic arms, Dehaye illustrated how complex multi-stage operations in uncertain environments mirror modern cyber threats.

Core Innovation: Exo-Language for Machine Communication

SGDL has developed a proprietary « exo-language » designed for direct machine-to-machine communication: « We have machines using a language to generate models that will then be used further by machines »

This breakthrough enables AI systems to model both physical objects and their interaction logic simultaneously—technology currently running on the International Space Station and Mars rovers.

Space-to-Security Transfer

The company is transferring this space-proven technology to cybersecurity applications:

  • AWS Policy Modeling: Physics-based interaction analysis for cloud security

  • Autonomous Threat Response: Multi-stage defensive systems matching attacker sophistication

  • Space-Filling Vector Optimization: Mathematical curves for efficient data navigation

The IKEA Connection

Dehaye's unexpected analogy: IKEA's Lausanne store design mirrors SGDL's mathematical approach. The store's path « crawls through space, a maximal amount of space » - exactly how SGDL's space-filling curves optimize vector spaces for AI reasoning.

Strategic Advantage

Unlike traditional AI security relying on pattern recognition, SGDL models fundamental interaction principles. This physics-first approach runs efficiently on basic hardware while delivering space-grade reliability—crucial for edge computing and resource-constrained environments.

European AI Leadership: SGDL's Panoramai Award highlights their contribution to European AI competitiveness through mathematical rigor rather than data scale.

Paul-Olivier Dehaye is the CEO and Founder of Hestia.ai, a Swiss company established in 2021 that specializes in creating sustainable, transparent, and ethical AI solutions for businesses. A mathematician by training, Dehaye holds a PhD from Stanford University and has held fellowships at Oxford and ETH Zurich. He is widely recognized for his pivotal role in uncovering the Facebook-Cambridge Analytica scandal, which exposed the misuse of personal data for political influence