Question

A property insurance company wants to build a solution that lets adjusters upload photos of vehicle damage along with a short text note, and receive a written assessment describing the visible damage and likely repair categories. The leadership team is evaluating which type of generative AI capability best fits this need. Which capability should they prioritize when selecting a model?

Accepted Answer

A multimodal model that can accept both image and text inputs and generate natural-language output

Answer

A text-only large language model paired with a rule-based keyword filter

Answer

A traditional image-classification model that outputs a single damage label per photo

Answer

A speech-to-text model that transcribes the adjuster's spoken notes into the report