An enterprise is considering deploying an AI solution that will be used across multiple business...

The CAIPM framework emphasizes selecting AI architectures that maximize scalability, reuse, and long-term value across enterprise functions. The scenario clearly describes an approach where a single, shared core model is leveraged across multiple domains, with domain-specific customization layered on top. This is the defining characteristic of Foundation Models.

Foundation models are large, pre-trained models built on broad datasets and designed to serve as a general-purpose base. They can be adapted to various use cases—such as customer service, content generation, analytics, or internal knowledge systems—through fine-tuning, prompting, or lightweight customization. This approach avoids building multiple isolated models, reducing development cost and improving consistency across the organization.

Option B (Generative AI) refers to a capability (content creation) rather than an architectural strategy. Option C (Machine Learning) is too broad and does not capture the shared-core design principle. Option D (Large Language Models) is a subset of foundation models focused specifically on language tasks, but the question emphasizes strategic reuse across domains, not just language specialization.

CAIPM highlights foundation models as a key enabler of enterprise AI strategy because they support modular scaling, faster deployment of new use cases, and alignment with long-term investment priorities.

Therefore, the correct answer is Foundation Models, as it best reflects a shared core capability with domain-specific adaptations across the enterprise.