Taboola Launches Ad Network for AI Chatbots and LLM Apps

Taboola opens ad marketplace to AI platforms

Taboola has launched an advertising network purpose-built for large language models, chatbots, and virtual assistants, extending the monetization infrastructure behind its DeeperDive product to external AI platforms.

The new offering allows third-party conversational AI applications to tap into Taboola's base of approximately 20,000 performance advertisers, inserting native sponsored content directly into AI-generated responses. Taboola will handle ad sourcing, targeting, and billing while sharing revenue with AI platform partners under its standard marketplace model—typically retaining about 35% of ad spend and distributing 65% to partners, according to CEO Adam Singolda.

Why it matters

As AI companies struggle with the high costs of running LLMs at scale and face consumer resistance to multiple subscriptions, advertising emerges as a critical revenue path. Taboola's move creates infrastructure for monetizing conversational interfaces before AI platforms build their own ad stacks, potentially establishing early standards for how ads appear in agent-driven experiences. The company's positioning as an alternative to Google and Facebook for open-web publishers now extends into the AI layer, where intent signals from queries could rival traditional search advertising.

How the monetization engine works

DeeperDive functions as an answer engine embedded on publisher pages, transforming editorial content into interactive chat experiences. Users can ask follow-up questions or request summaries, with responses generated through a combination of LLMs, retrieval systems, and Taboola's proprietary intent graph.

The platform inserts native ad units that visually match AI-generated responses. When a user asks about purchasing a home, for instance, the system can surface mortgage or home-finance offers as sponsored results within the conversation thread. Singolda noted that DeeperDive already delivers the highest conversion rates across Taboola's entire network.

Currently, the system relies heavily on Taboola's first-party audience data, collected through integrations across thousands of publisher sites. The company plans to shift toward contextual and query-level intent signals, making ads function more like search units triggered by real-time user questions.

Positioning against walled gardens

Singolda framed the LLM ad network as part of Taboola's broader strategy to become "the economic layer for anyone outside of Google and Facebook." The company's direct integrations with roughly 9,000 publishers feed an intent graph tracking user interests across the open web—from sports preferences to mortgage shopping.

This cross-site visibility, Singolda told Digiday, gives Taboola an advantage over general-purpose LLM providers that lack insight into users' broader browsing behavior. As AI agents increasingly handle consumer journeys from discovery to purchase, connecting conversational intent to longitudinal interest data becomes strategically valuable.

Publishers currently using DeeperDive are seeing monetization increases of approximately 10% on top of existing Taboola revenues, though the company is not offering minimum revenue guarantees. The business operates on a cost-per-click basis, mirroring Taboola's core content recommendation marketplace.

These details were first reported by Digiday.