Generative Player for LLM Applications
Octo's player components callable as tools by large language model applications and rendered inline, so your video, products, and community appear inside AI assistants under your rules.
The generative player brings your Genuin player into AI assistants. A large language model (LLM) application calls one of Octo's player components as a tool, and the assistant renders it inline. Your brand's video, products, and community then appear under the same rules as onsite.
This capability arrives as a proposal when your network needs it, filled in from your brand context, and goes live when you confirm it. It's part of Octo, and Octo: the orchestration and generative commerce layer explains where Octo appears and how it's governed.
When to use it
It's in early access, for brands that want presence in assistants. You choose what appears, and your existing rules govern it.
It's different from AdCP: make your inventory discoverable to AI buyer agents. AdCP lets AI buyer agents discover and buy your packages. The generative player puts your content in front of viewers inside the assistants they use.
| Surface | Who it serves | What appears |
|---|---|---|
| Generative player | Viewers inside AI assistants | Your video, products, and community |
| AdCP | AI buyer agents planning media | Your packages, as products to buy |
| Your own pages and apps | Viewers on your properties | Your placements, through the SDKs |
How it works
- An LLM application calls one of Octo's player components as a tool. The assistant renders the component inline.
- You decide which groups and products can render, and in which assistants.
- The same brand safety and consent rules apply as onsite. Octo never acts outside your brand safety profile, set in Set moderation, roles, and brand safety.
- Viewer data follows your agreement and the privacy controls in Privacy and VPPA.
- Paid content always carries a disclosure label, as Paid Partner Posts and Paid Partner Takeovers describes.
Set it up
Octo proposes the generative player with these steps already done. You review the result and confirm it.
- Ask your customer success manager (CSM) for early access.
- Define which groups and products can render. Brand communities and groups: structure content for relevance explains how groups are organized.
- Confirm the rules. The same brand safety and consent rules apply as on your own placements.
Settings
| Setting | What it controls |
|---|---|
| Allowed content | Which groups and products can render inside assistants |
| Destinations | Which assistants can render your content |
| Disclosure | The disclosure shown with rendered content |
Brand safety and consent aren't separate settings here. They follow the profile and consent rules you already set for your own placements.
What it reports
Renders are reported by assistant, next to their engagement and outcomes. Measure engagement: watch time, lift, and cohorts explains how engagement is counted.
- Renders by assistant: how often each assistant rendered your content.
- Engagement: what viewers did with the rendered content.
- Outcomes: the clicks, purchases, and other results those renders led to.
Build on it
The same player renders on your own site today with Web SDK 2.0.6. Embed the Genuin Web SDK on your website covers loading the script and your first embed. Add a Genuin placement to your website renders a placement whose videos follow each page's context. Customize the Genuin video player sets sizes, tile layouts, and the profile menu.
The feed and resolve calls in the Business Account API are proposed, not live. Live API today and the Business Account API roadmap maps what you can call now to what's planned.
