1 Why model adaptation?
This chapter covers
- Questions every team asks before adapting an LLM
- Scenarios where adaptation pays off and the ROI math behind each
- Enterprise pain points: accuracy gaps, API cost at scale, regulatory constraints, vendor lock-in
- When NOT to adapt: the cases where prompting alone is the right answer
- A decision framework for choosing your first technique
- The base model, dataset, and tools we use for hands-on examples
A demo of the new policy assistant on Thursday nailed every test case. Leadership pushed for users by Friday. By the following Wednesday, the support queue caught the first fabrication: a paragraph the model invented, citing a reimbursement clause that doesn't exist and recommending escalation to a manager who left two years ago. The phrasing was plausible enough that a reviewer had to know the policy by heart to flag it. By then, four other employees had already acted on similar invented guidance elsewhere in the assistant.
That failure is what happens when a general-purpose large language model (LLM) meets an organization for the first time. The model is fluent and broadly capable, but it does not know your terminology, workflows, or current policy, and it has no way to flag when it is guessing. This book closes that gap and teaches you when not to try.