Generative AI Leader Practice Question: A data scientist wants to generate photorealistic…

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Veo

  Why it's wrong here

  Veo generates video from text or image prompts, so it cannot output the still product photographs the catalogue requires. It is tempting because Veo produces realistic visual content, but that capability targets motion; text-to-image generation for static, brand-consistent product shots needs Imagen instead.

• ✓

  Imagen

  Why this is correct

  Imagen on Vertex AI generates photorealistic images from text prompts, directly satisfying the catalogue requirement. Its controls for brand consistency and product fidelity reduce distorted features, unlike general-purpose or conversational models. This makes it the appropriate Google Cloud service for producing reliable e-commerce product imagery at scale.

• ✗

  Chirp

  Why it's wrong here

  Chirp is a speech model for transcription and text-to-speech, producing audio rather than images, so it cannot generate product photographs. It is tempting because Chirp is a Google Cloud generative AI service, but its modality is voice; photorealistic text-to-image generation for a catalogue requires Imagen.

• ✗

  Gemini Pro Vision

  Why it's wrong here

  Gemini Pro Vision analyses and describes existing images, so it cannot synthesise new photorealistic product photographs from text prompts. It is tempting because it handles visual content, but that is understanding rather than generation; producing brand-consistent catalogue images requires Imagen's text-to-image capability.

Go deeper

Related to this question