Your AI Reality Check
Hosted by Ant Open Source and Ant Ling
As part of Open Source AI Week 2026, this one-night discussion brings together builders and researchers to explore what open-source AI looks like when put into real-world.
We will feature two focused panels:
Can We Trust What the Benchmark Says?
Can an Agent System Be Smarter Than Its Best Model?
Together, we’ll also examine how domain-specific models are being developed for healthcare and finance, how benchmarks translate into real-world impact, and whether agent systems can go beyond the capabilities of their underlying models.
Drawing on Ant Group’s experience building financial technology and digital payment infrastructure at global scale, we’ll bring a practical perspective to the conversation—but the focus is broader: how can the open-source community build AI that is more reliable, accessible, and useful across high-stakes industries?
This event is designed to be more than a series of talks.
Join us for high-signal, 1-1 conversations with people building, researching, and deploying AI. We hope to create a space where thoughtful questions lead to meaningful connections, new collaborations, and the next generation of open-source ideas.