Enterprise architecture has always been about making deliberate choices — which systems to integrate, which standards to enforce, which capabilities to build versus buy. AI introduces a new layer of complexity: systems that learn, adapt, and generate outputs that are probabilistic rather than deterministic.
What changes for architects
Traditional architecture assumes predictable system behaviour. AI systems require architects to think about data pipelines, model governance, prompt management, and human-in-the-loop workflows as first-class architectural concerns — not afterthoughts bolted onto existing platforms.
Practical implications
- Data architecture becomes the foundation for every AI initiative
- Integration patterns must account for non-deterministic outputs
- Governance frameworks need to cover model lifecycle, not just software lifecycle
- Security models must address prompt injection, data leakage, and model access
The organisations that will succeed are those that treat AI as an architectural capability — not a feature to be added to individual applications in isolation.