Snap is testing a new ad format that places sponsored content directly inside chat conversations, delivered by AI agents built to sound like part of an actual exchange.
The move is a shift away from traditional display ads - banners, overlays, sponsored stories - toward ads embedded in conversational flows. Instead of an ad appearing between messages, the AI agent itself becomes the delivery vehicle, initiating or responding within chat in a way that mimics normal back-and-forth messaging.
This builds on Snap's existing "My AI" chatbot, which already sits in the chat section of the app. By routing sponsored content through a conversational agent rather than a separate ad slot, Snap is betting users engage more when an ad doesn't immediately read as one.
For advertisers, the pitch is straightforward: higher engagement rates than static placements. For users, the tradeoff is murkier. When an AI agent is both a helpful assistant and an ad delivery mechanism in the same interface, the line between genuine assistance and paid content gets intentionally blurry. Snap has indicated it will label sponsored interactions, but "labeled" and "designed to feel like conversation" are in tension by design.
Who's Already Doing This
Snap isn't operating alone here. Meta has tested AI assistants inside WhatsApp and Messenger, and brands have run chatbot-style ad experiences on various messaging platforms for years. What's different about Snap's approach is embedding paid content into a general-purpose AI agent inside a consumer social app - not a branded bot the user opted into, but the default AI assistant the app pushes on everyone.
The broader pattern is clear: as users spend more time in chat interfaces and less time scrolling feeds, ad budgets follow attention. AI-powered chat is where that attention is moving, and Snap wants to be first among social apps to have a monetization model built around it rather than bolted on.
The real test isn't whether advertisers buy in - they will. It's whether users accept being sold to by the same agent they ask for restaurant recommendations.