Personalised Pricing: The Demise of the Fixed Price? (arxiv.org)

Researchers review how online sellers can technically offer individualized prices using consumer data and algorithms, and they pair that capability with new survey evidence showing broad consumer mistrust: most respondents view personalized pricing as unfair, want to be informed when it’s used, and many favour bans on the practice. The chapter argues that the most controversial forms of price discrimination—those driven by profiling and automated decision-making—fall within the scope of the EU’s GDPR, which not only demands transparency but, for such uses, may require prior informed consent and protections against opaque automated decisions. Despite these legal obligations, industry practice rarely reflects disclosure or consent mechanisms. For the AI/ML community this matters because personalized pricing is an application of profiling, recommender systems and dynamic-pricing algorithms that directly ties model outputs to economic outcomes for individuals. Practitioners should expect stronger regulatory and public scrutiny: lawful bases for processing, transparency, refusal/opt-out options, DPIAs, and auditability will become essential. Technically, teams will need interpretable models, logging for accountability, fairness-aware optimization (or group-aware pricing constraints), and privacy-preserving approaches to reduce discrimination risk. The gap between legal requirements and current industry behaviour points to imminent enforcement, reputational risk, and a practical need for guardrails around algorithmic pricing.