OpenAI's move into shopping features is hitting a major hurdle: sales tax complexity. This forces them to build a sophisticated backend for tax collection and remittance, similar to Amazon, rather than just being a thin AI layer. This will necessitate a hiring push for compliance and tax experts.
Related Insights
The primary obstacle for OpenAI's shopping features isn't the transaction layer, but the complex task of standardizing inconsistent product data (sizing, pricing, inventory) across millions of merchants. This foundational data problem requires deep collaboration with partners and explains the slow, deliberate rollout.
For OpenAI's commerce features to succeed, it's not enough to build one-click checkout. They must fundamentally retrain hundreds of millions of users to trust a new purchasing workflow inside a chatbot, breaking deeply ingrained habits of searching on ChatGPT then buying on Google or Amazon.
While a 4% fee seems reasonable for new customer acquisition, it becomes a burden if users who discovered a product organically then use ChatGPT for checkout convenience. This dynamic forces merchants to pay OpenAI for customers they didn't acquire through the platform.
By integrating shopping into ChatGPT, OpenAI can become a massive e-commerce engine. With a potential take rate of 15-30%, similar to Amazon or Meta, capturing just 20% of the $1.2T U.S. e-commerce market would generate tens of billions in new, high-margin revenue.
OpenAI's 4% fee for in-app purchases creates a risk for merchants. If consumers start using ChatGPT as their primary purchasing interface, it could intercept sales that originated from a brand's own marketing. A customer might see a product elsewhere, then buy it via ChatGPT, imposing a new tax on an otherwise organic conversion.
OpenAI's path to profitability isn't just selling subscriptions. The strategy is to create a "team of helpers" within ChatGPT to replace expensive human services. The bet is that users will pay significantly for an AI that can act as their personal shopper, travel agent, and financial advisor, unlocking massive new markets.
The biggest hurdle for AI shopping agents isn't the AI, but the messy reality of retail logistics like product data and sales tax. While OpenAI focuses on the AI layer, Amazon's true advantage is its deeply entrenched commerce infrastructure, which is far harder for competitors to replicate.
Despite powerful new models, enterprises struggle to integrate them. OpenAI is hiring hundreds of 'forward-deployed engineers' to help corporations customize models and automate tasks. This highlights that human expertise is still critical for unlocking the business value of advanced AI, creating a new wave of high-skill jobs.
Amazon's potential commerce partnership with OpenAI is fraught with risk. Allowing ChatGPT to become the starting point for product searches threatens Amazon's highly profitable on-site advertising revenue, even if Amazon gains referral traffic. It's a classic battle to avoid being aggregated by another platform.
OpenAI is more public and aggressive with its shopping features (partnering with Shopify, DoorDash) than its ad strategy. By first attracting thousands of merchants to its e-commerce waitlist, it's establishing a foundational transaction layer. This de-risks its future ad platform by ensuring a ready base of paying customers.