# How to choose an AI project - Set of things AI can do (need AI experts) - Needs technical diligence to determine what is actually feasible - Can an AI system meet the desired performance - How much data is needed - Required resources and engineering timeline - Set of things that are valuable for your business (need domain/business experts) - Needs business diligence - Could lower costs? - Could increase performance? - Could increase revenue? - Could help launch new products/businesses? - Create financial/economic models - Find ideas that are at the intersection of those sets - Ethical diligence - Would this make humanity better off? Framework for brainstorming interesting projects to explore: - Think about automating tasks rather than automating jobs - Focus on tasks and processes first - What are the main drivers of business value? - What are the main pain points of the business? It's possible to make progress even without tons of data. Even if more data is generally better, less data can still lead to valuable results.