Retail has been running recommendation and forecasting models for years. What changed is that customers increasingly arrive through AI assistants, sometimes with an agent completing the purchase, and that pricing and review practices driven by algorithms now draw specific laws and enforcement. This course covers both the established toolkit and the new channel.
You will cover semantic product search and LLM catalog enrichment, two-tower retrieval and ranking for recommendations, honest experimentation, how AI shopping agents discover products through feeds and structured data, the Agentic Commerce Protocol from OpenAI and Stripe, Google's Universal Commerce Protocol, and the Agent Payments Protocol (AP2) that Google created and handed to the FIDO Alliance in 2026, fraud and chargeback questions when an agent is the buyer, forecasting and markdown optimization, the New York Algorithmic Pricing Disclosure Act and antitrust risk from shared pricing software, uplift modeling for promotions, generative product copy and imagery, returns and order-status agents, shelf monitoring and loss prevention, the FTC rule on fake reviews and testimonials, and biometric privacy in stores. Protocol and legal status are stated as of October 2026 and change quickly; the course is educational, not legal advice.