The Liverpool Institute for Boys was a well-regarded grammar school in the north of England in the 1950s and 1960s. It provided rigorous secondary education to bright working-class boys who had passed the eleven-plus examination. Its music curriculum was, by the standards of the time, competent and thorough.
Two of its students in that period were Paul McCartney and George Harrison.
The school’s music teacher, by multiple accounts, did not identify either of them as having exceptional musical talent. They received standard music instruction. They were not encouraged toward musical careers. They were not recommended for scholarships or specialized training. The teacher, working within the evaluation frameworks available to him, saw two ordinary boys with ordinary aptitude.
What McCartney and Harrison went on to build with John Lennon and Ringo Starr is the most commercially and critically successful body of popular music in the 20th century. The failure of the Liverpool Institute music teacher to recognize their talent is not a story about one teacher. It is a story about how evaluation systems structurally fail to identify certain kinds of talent.
The Elvis parallel
The same story exists in many other cases. Elvis Presley attended Milam Junior High and later Humes High School in Memphis. In Tupelo, before moving to Memphis, the eighth-grade glee club teacher reportedly told his mother that Elvis did not have singing talent and should not be encouraged in that direction.
The teacher was applying the evaluation framework available to her. That framework valued specific vocal characteristics — trained tone production, controlled vibrato, formal technique. Elvis’s talent was of a fundamentally different kind. It was not visible within the framework the teacher was using.
Sir Ken Robinson’s framing
Sir Ken Robinson used these stories in his talks on education and human potential to make a specific point.
“Human resources are like natural resources. They are often buried deep. You have to go looking for them, and you have to create the conditions in which they can express themselves.”
The insight is not that the teachers were incompetent. The insight is that evaluation systems, by their nature, evaluate against known patterns. Talent that fits the known patterns gets identified. Talent that does not fit the known patterns is structurally invisible to the evaluation system.
The 2026 application
Every senior leader I work with in 2026 is running an organization whose talent evaluation systems were calibrated to identify pre-AI knowledge worker capabilities.
These systems evaluate specific things well. Analytical rigor. Written communication. Presentation skills. Domain expertise. Team leadership. Business acumen. Every one of these is a legitimate capability that continues to matter in AI-transformed work.
These systems evaluate other things poorly. Some of these poorly-evaluated capabilities are precisely the capabilities that determine who thrives in AI-transformed work.
The ability to formulate specific questions that produce useful outputs from AI models. The ability to construct workflows that combine AI capabilities with human judgment in non-obvious ways. The ability to identify AI outputs that are subtly wrong even when they appear plausible. The ability to iterate through many AI-generated options quickly, filtering aggressively.
None of these capabilities is a category on standard performance reviews. None of them appears on standard job descriptions. None of them is what senior managers were themselves selected against.
Which means the McCartneys and Harrisons of AI-native capability are sitting inside every enterprise, structurally invisible to the evaluation systems that determine promotion, compensation, and strategic responsibility.
Three practical questions
One: what specific AI-native capabilities are missing from your current evaluation framework? The list above is a starting point, not a complete inventory. What would you add?
Two: who in your organization is currently exhibiting AI-native capability without receiving proportional recognition? They exist. They are not being credited. In three years, they will either be your competitor’s most valuable hires, or they will have quietly become the load-bearing infrastructure of your organization while remaining under-titled and under-paid.
Three: what would it cost to build a parallel evaluation track for AI-native capability? Not to replace the existing evaluation system. To supplement it. Almost every enterprise could implement this cheaply. Almost none are doing it.
The closing thought
The Liverpool Institute music teacher was not stupid. He was competent. He was working within the evaluation framework available to him. Within that framework, McCartney and Harrison were ordinary boys with ordinary aptitude.
The framework was the problem. Not the teacher.
In 2026, the frameworks most enterprises use to evaluate talent are calibrated to a pre-AI knowledge worker world. Those frameworks are systematically invisible to AI-native capability. The most valuable people in your organization, three years from now, will be people whose current evaluations describe them as ordinary.
Some will leave. Some will stay. In either case, if you had recognized them earlier, you would have compounded their capability inside your organization instead of losing it or under-utilizing it.
The world has changed. The leaders who notice will be the ones the next decade is built around.
