AI agents are accelerating software delivery. Code ships faster than ever before. The open question is whether architecture improves at the same pace.
Developers have always made implicit architectural decisions. What has changed is the speed at which those decisions are made — and increasingly, delegated. When AI systems generate code based on incomplete context, assumptions are embedded into design choices as if they were intentional. These assumptions compound over time, often without visibility.
This session explores how implicit architectural drift emerges in AI-assisted development, how cognitive biases amplify the problem, and how teams can retain architectural clarity and deliberate design while leveraging AI acceleration.
The focus is on keeping architecture intentional, explicit, and human-led in an AI-accelerated engineering environment.