Fine-tuning

Further training a model on your own examples so a style or task becomes default behaviour, versus steering it per-request with prompts and context. Most marketing teams never need it: prompting with good context is cheaper, faster to change, and usually sufficient.

Evidence from the library

Framing content around how people actually prompt - by product or by context - can win more model visibility.Principle Over-optimizing creative for AI models risks a second crisis of creative effectiveness.Warning Marketing mix modelling has grown too complex for human judgment alone - the useful AI layer is a plain-English front end for CMOs to ask what to change next quarter.Principle People love disruption in content while models want direction and directional cues instead.Principle Brands should not over-optimize for machines at the expense of humans, since people - not agents - ultimately make the purchase, and brand building remains essential even in agent-heavy categories.Warning

Source

Marketing Definitions

standard AI practice