Premium membership
Continue this concept with Premium
Premium includes all published courses, graded answers, concepts and PDF guides. See the price and access terms on the pricing page.
View Premium accessone payment · six months of site access
Advanced
Platform Risk and Lock-In: Depending on Model Providers Without Being Trapped
AI applications depend on model providers that can change prices, deprecate models, alter terms or enter the application's market. This guide covers the forms platform risk takes, a ranking of lock-in by switching cost (evaluations, data, tool contracts, prompts, SDK), mitigation architecture, contract terms, and how to respond when a provider competes with you.
CONCEPT / STRATEGY & MOATS
RELATED CONCEPTS
PRACTISE THIS CONCEPT
AI Strategy, Moats & PricingThe model provider we build on just launched a feature that does our core use case, for free inside their assistant. What do you do?→AI Strategy, Moats & PricingA major model provider offers discounts, early access and co-marketing in exchange for exclusivity. Do we take it?→AI Strategy, Moats & PricingFor a new invoice-processing feature, should we use a commercial model API, fine-tune an open model, or build our own? How do you decide?→AI Strategy, Moats & PricingOur CEO wants us to train our own foundation model to 'own our AI future'. We're a 2,000-person software company. What do you recommend?→AI Strategy, Moats & PricingYou're a product leader at a model lab. Should the lab release its next model's weights openly? Make the case either way.→Agentic ProductsShould we build our agents on a model provider's hosted agent platform, or own the orchestration ourselves?→