World Congress 2026 Europe

Beyond LLM Agents: A World Model for Visual Software Testing

July 9, 2026 14:15 – 14:20 Β· 5 min Airstream 1

What this session covers

What if a model could learn software the way humans do, through pixels, mouse, and keyboard alone? This talk introduces the Vision Action Model, a foundation model that understands UI concepts, not coordinates, and executes them deterministically. Purpose-built for software testing. Trained purely on interaction data.

World Congress 2026 Europe

Related talks

Open session

Giving AI eyes: How to build a dashboard you can't see

  • Josh Hobson

    AG Grid

    Senior Engineer

Open session

Fine-Tuning Small Language Models for Agentic AI

  • BjΓΆrn Buchhold

    CID

    Technology Evangelist

Open session

Testing AI Agents: Automated Evaluation for Chatbots & RAG Systems

  • Sebastian Messingfeld

    Eurowings Digital

    Staff Engineer

Open session

Are Classical Automation Frameworks Dead? How AI Agents Are Transforming QA

  • Andrei Nutas

    Atos

    Technical Test Architect

All sessions at this congress