The AI Value Canvas is designed to support decisions beyond a particular employer's AI budget or a single project. It is a way of thinking through whether an AI initiative deserves your time, your team's trust, and your organization's money — questions that remain relevant as roles, teams, and initiatives change.
The next time someone pitches you an AI idea — or you're the one pitching it — the useful move isn't to ask whether the technology works. It's to ask whether users actually want it, whether AI is genuinely the right tool for the job, whether it can be built and run responsibly, and whether the economics hold up once you account for the parts that are easy to forget. Four questions, in that order, before a line of code gets written.
Design AI With Human Support and Accountability
The aim is not technology for its own sake. Use AI where it is genuinely helpful, but do not treat it as a substitute for human judgment, responsibility, or care. Where the Canvas supports AI for a bounded task, make the human handoff, accountable owner, and escalation path clear before launch.
Spain's Ley 10/2025 offers one jurisdiction-specific example. For businesses within its scope, automated systems cannot be the only customer-service channel: people must be able to request personalised assistance.
The Air Canada chatbot case offers a related accountability lesson. When a chatbot gave incorrect bereavement-fare information, the organisation remained responsible for the outcome. The point is not to avoid AI; it is to use it wisely, with reliable information, human support for consequential cases, and a clear way to correct things when they go wrong.
Keep the Canvas. Use it on the next idea, and the one after that. It gets more useful the more places you apply it.