Definition
What is generative ai?
Generative AI describes models that produce new content — text, structured data, code, images or audio — based on patterns learned from training data, rather than selecting from predefined options. In business use it most often means large language models applied to documents, correspondence and structured extraction.
Updated 3 September 2026
Why it matters
- It made unstructured text and documents economically processable, which is where most administrative cost sits.
- Output quality varies by input, so evaluation and review design become part of the operating model.
- Usage-based pricing means cost scales with adoption rather than being fixed at build time.
Business example
A finance team uses a model to extract line items from supplier invoices with varied layouts, then validates values against master data and routes low-confidence items to a reviewer.
Common misconceptions
- 'Generative AI is a search engine.' It generates plausible text; grounding it in your data is a design decision.
- 'Accuracy is a property of the model.' It is a property of the model plus the data, the prompt, the validation and the review process.
- 'Costs are trivial.' Inference is cheap per call and material at volume, and review time usually costs more than inference.
Related concepts
- Large language model
- Retrieval-augmented generation
- AI automation
Where would this apply in your business?
Merjora maps your workflows and quantifies the opportunities worth acting on.
Discover your AI opportunitiesRelated reading
Editorial standard. Merjora publishes analysis, frameworks and publicly documented examples. We do not publish invented statistics, unattributed benchmarks or unverified customer stories.