Pinterest launches Ask Pinterest, a conversational AI shopping app to compete with search-based discovery

According to TechCrunch AI, Pinterest launched Ask Pinterest on June 17, an experimental standalone application that replaces visual browsing with a conversational interface for product discovery. The app allows users to ask natural-language questions to receive personalized shopping recommendations, supporting multi-step workflows like dinner planning or room furnishing. The company plans to incorporate learnings from Ask Pinterest into its flagship app.

Conversational AI as a standalone experiment

Pinterest is isolating the conversational shopping experience in a separate app rather than integrating it directly into its main platform. This architectural choice lets the company test AI features without risking disruption to an existing user base that relies on visual discovery. According to TechCrunch AI, the app handles complex, multi-step queries that don’t fit traditional pin-based search—scenarios where users benefit from maintaining context across multiple interactions.

The limited-access rollout mirrors how other major platforms (Google, Meta, Shopify) have approached agentic shopping features. By operating Ask Pinterest independently, the company gains flexibility to iterate on the conversational interface and measurement approaches before deciding which elements are ready for broader integration.

Leveraging the Taste Graph for personalization

Ask Pinterest builds on Pinterest’s proprietary Taste Graph, an internal data structure that maps user interests and aesthetic preferences. The app augments this company-level signal with individual user context—saved Pins and Boards—to deliver recommendations tailored to each query. This dual-layer personalization strategy allows Pinterest to compete with general-purpose AI shopping assistants by injecting domain-specific taste data that generic chatbots cannot access.

The Taste Graph represents a defensible moat against licensing deals where competitors might extract value from Pinterest’s discovery algorithms. According to TechCrunch AI, Pinterest has prioritized using its own data to train models rather than licensing recommendations to other AI services.

Broader adtech AI initiatives

Ask Pinterest is part of a larger push into advertiser-focused AI tools. Pinterest also announced Pinterest Model Context Protocol (MCP), an infrastructure layer for advertisers managing campaigns on the platform, along with a beta AI assistant in Ads Manager and Performance+ creative, a global feature that helps advertisers select ad creatives likely to maximize performance. These tools signal Pinterest’s intent to embed AI throughout both consumer and advertiser workflows ahead of Cannes Lions 2026, the industry’s main advertising conference.

Why This Matters

Ask Pinterest signals a fundamental shift in how visual-discovery platforms compete with search engines and chatbots for shopping intent. If the conversational interface proves faster and more intuitive than visual search for certain use cases, Pinterest will have validated a pathway to recapture users who might otherwise default to ChatGPT or Google Shopping for product decisions. The Taste Graph becomes a strategic advantage only if Pinterest can surface recommendations that generic LLMs cannot—a bet that depends on converting experimental engagement into retention metrics. For advertisers, the MCP and Performance+ tools suggest that Pinterest is moving beyond impression-counting toward outcome-oriented marketing, which could justify higher CPMs if the company can demonstrate campaign-level ROI improvements.