The Discovery ProblemmiguelPostsAbout2026-09-03The Discovery ProblemThe biggest bottleneck to AI adoption is a simple question: what can this do for me?The hard part is that you don’t know what you don’t know. You don’t know what a button does until you press it. Press meYou don’t know what a prompt can produce until you write it, hit go, and watch it run.As long as capabilities stay locked behind a blank text box, the possibilities stay invisible. That’s a discovery problem, and it’s the one we’re still stuck on.There are partial fixes. Templates give people something to run without needing to invent the request themselves. But then the question becomes relevance. Do these templates actually match your work? Do you care? Context helps too: a system that knows about you can suggest things that matter to you instead of things that matter in general. Both help. Neither solves it.Alan Kay has a metaphor for this. Imagine you’re an ant at the bottom of the Grand Canyon. You look up, and your entire notion of the sky is a thin sliver of blue between two canyon walls. Someone standing on the rim sees the whole blue plane. Same sky, completely different sense of what exists. It’s not that the ant is less capable. It just can’t see the axis of possibility from where it’s standing.Watch Alan Kay explain itThat’s the gap between a skilled AI user and everyone else. Take a non-technical marketing person and someone fluent in agents and tool use. The agent-fluent person can watch the marketer work for an hour and immediately see a dozen things to automate, delegate, or reinvent, including things the marketer hasn’t even tried yet. But put the most intelligent tool in the world in front of the marketer, and they’re staring at a blank prompt, unsure what to type. All that intelligence, and no way to see it.This is the strange state we’re in: the system could do almost anything, but it requires the user to already know what to ask for. Too much of the work of discovering what’s possible falls on the person, when it should fall on the system. You’d expect something this advanced to reveal its own capabilities — gradually, contextually, in ways that match your actual work. We’re not there yet.Somehow, the interface has to start showing you the sky.← All posts

The primary obstacle to widespread adoption of artificial intelligence lies in the fundamental question of utility: what can this technology do for the individual user. This difficulty stems from an inherent lack of knowledge, as users do not know what they do not know; capabilities remain inaccessible because all potential functions are locked behind a blank input interface. This situation defines the discovery problem, which remains a significant hurdle in the development of practical AI applications. While partial solutions exist, such as providing templates, these methods face challenges regarding relevance, as they may not align with the user's specific professional context or needs. Furthermore, incorporating context, such as building a system that knows about the user, offers some benefit by suggesting relevant information rather than general possibilities, yet neither template-based suggestions nor simple contextual awareness fully resolve the core issue of discovery.

The disparity in understanding can be illustrated by Alan Kay’s metaphor concerning an ant at the bottom of the Grand Canyon viewing the sky as a thin sliver, contrasting with someone on the rim who perceives the entire plane. This analogy highlights the gap between advanced technical fluency and practical application. An individual proficient in agent usage and tool manipulation possesses the ability to perceive numerous automation opportunities when observing another person working, yet placing the most intelligent tools before those who lack domain expertise forces them to confront an empty prompt without guidance. This creates a paradoxical state where immense system intelligence is constrained because the user must already possess the knowledge required to formulate effective requests.

The underlying tension is that the burden of discovering what is possible is currently placed on the user rather than the system itself. Such advanced systems should ideally reveal their capabilities incrementally and contextually, mirroring the user's actual work. The current dynamic prevents this self-discovery, resulting in a situation where the potential of the technology remains obscured. Therefore, for AI to be truly effective and accessible, the interface must evolve to actively guide the user toward understanding these possibilities, effectively showing them the full scope of what is achievable.