“What’s the problem you’re trying to solve?” Clayton Christensen, the late Harvard business professor, was famous for posing this aphoristic question to aspiring entrepreneurs.

By asking it, he was teaching those in earshot an important lesson: Innovation, alone, isn’t the end goal. To succeed, ideas and products must address fundamental human problems.

This is especially true in healthcare, where artificial intelligence is fueling the hopes of an industry desperate for better solutions.

But here’s the problem: Tech companies too often set out to create AI innovations they can sell, rather than trying to understand the problems doctors and patients need solved. At many traditional med-tech conferences and trade shows, for example, talks and sales pitches focus squarely on the technology while routinely overlooking the human fears and frustrations that AI can address.

Because of this failure to prioritize human needs above business interests, medicine’s most-hyped AI applications have, repeatedly, failed to move the needle on public health, patient safety or healthcare costs.