OpenAI Begins Testing Native Ads in ChatGPT
OpenAI is piloting native advertising within ChatGPT, marking a significant step toward monetizing its popular AI assistant. The first test involves partnerships with major retail brands including Target, Adobe, and Williams-Sonoma. The experiment is being closely watched as a potential new business model for AI-driven information and curation platforms.
- The move is driven by immense financial pressure; despite a projected $20 billion in 2025 revenue, OpenAI has committed over $1.4 trillion to AI infrastructure and only about 5% of its 800 million weekly users pay for a subscription. - Ads will only appear for users on the free tier and a new $8/month "ChatGPT Go" plan, while Plus, Pro, and Enterprise subscribers will remain ad-free. - The ad system uses contextual targeting based on the current conversation, not personal user data, which OpenAI has pledged not to sell to advertisers. Ads are also restricted from appearing in conversations about sensitive topics like health or politics. - Formats are designed to be native to the chat interface, including "Sponsored Recommendations" that appear below an answer, "Companion Display Ads" in a sidebar, and interactive "Native Content Cards" for products. - Wall Street analysts estimate the new advertising business could generate over $25 billion in annual revenue for OpenAI by 2030, potentially more than doubling the company's 2025 revenue. - The strategy creates a clear market division: competitor Anthropic is actively advertising its Claude chatbot as an ad-free alternative, a message that has resonated with users concerned about trust. - A recent Ipsos poll highlights potential user friction, with 63% of adults agreeing that ads in AI search results will decrease their trust in the platform's answers. - Major advertising holding companies, including WPP, Dentsu, and Omnicom, are participating in the pilot, coordinating campaigns for clients like Ford, Mazda, and HelloFresh.
