In One Sentence
Generative AI is artificial intelligence that, when a human says "make me X", automatically generates text, images, audio, video, and so on.
What does this look like in practice?
- Ask ChatGPT to "draft a business email" and it returns an appropriate piece of writing
- Tell DALL-E "an illustration of a cat swimming in space" and it generates an image
- Provide a synopsis to a video-generation AI (such as Sora) and it produces a short video
All of these are "Generative AI".
Why it matters (in a GEO context)
In the GEO context, we focus especially on text-generating AI (LLMs). Reasons:
- When a user asks "tell me about X", the AI generates the answer directly
- In generating the answer, the AI references information from the web and cites it
- If your company is cited, you enter the consideration set
In other words, with the rise of generative AI, "being cited in the AI's answer (GEO)" is becoming more important than "showing up on the first page of search results (SEO)".
Major generative AIs
| Domain | Representative Examples |
|---|---|
| Text | ChatGPT / Claude / Gemini |
| Image | DALL-E / Stable Diffusion / Midjourney |
| Video | Sora / Runway |
| Audio | ElevenLabs / Voice AI |
What GEO tactics target are primarily text-generating AIs (ChatGPT / Claude / Gemini). Image and video generative AIs are not direct GEO targets, but they are used in producing content-marketing assets.