What is a SuperSkill? Definition and the four tests

A SuperSkill is a capability that governs the other capabilities. It sits above any particular role, industry or tool, it survives the cycle that retires whatever you learned last, and it decides how well you use everything else you know. Seven of them are set out here, along with the four tests a capability has to pass before it counts as one. Rahim Hirji named the set on 22 October 2025, in "The Human + AI Era" for Box of Amazing, and developed it in SuperSkills (Kogan Page, 2026). The claim is to the framework and the naming; each of the seven words has its own literature and none of them is claimed.

Why the set exists#

Skills used to depreciate slowly enough to plan around. A person could train for a role, an organisation could plan a workforce, and the thing learned stayed useful long enough to pay for the learning. That is the arrangement that has broken, and the seven are an answer to the break rather than to the technology.

Three things about the break are measured rather than asserted, and they are the load-bearing ones.

The first is that the tasks disappearing are the ones people used to learn on. Junior work is where judgement was built. That is also the part most exposed to automation. That is the missing rungs. The problem therefore shows up as a pipeline problem years before it shows up as a capability problem.

The second is that the loss is not felt while it is happening. Developers using AI were 19 per cent slower and believed they were 20 per cent faster. Endoscopists' unassisted detection rate fell after exposure to an AI aid. In both cases performance held while the system was there, which is what makes decline hard to notice and easy to deny.

The third is that pairing a person with a machine does not reliably help. Across 106 experiments, human and AI pairs did worse on average than the better of the two alone. The gains went to the pairs where the person could tell which of them to trust on the question in front of them.

What the seven are for#

Each of the seven is a capability that decides how well everything else gets used, and each is named because the evidence for it existed before AI did. The set is a framework, not a finding. What sits under each name is a literature with its own measurements and its own limits, and each page says where those limits are.

There are two paths through what the tools make possible. One is drift, where the defaults decide and nobody chose. The other is design, where somebody decided in advance what stays human and wrote it down. The seven are the capabilities the second path runs on, and drift versus design is the argument for why the choice has to be made before the tools arrive rather than after.

What makes a SuperSkill#

The term skill has become so overused that it has lost precision. Job descriptions list dozens. Training catalogues offer hundreds. The implication is that capability is simply a matter of accumulation, that more skills means more value. That logic fails under conditions of rapid change.

Most skills are context-dependent. They work in specific roles, industries, or technological environments. When those contexts shift, the skills depreciate. A particular software proficiency, a specific process expertise, a narrow domain knowledge. These have value, but they do not compound. They erode.

SuperSkills operate differently. They are meta-capabilities that sit above roles, industries, and tools. They govern how someone learns new skills, adapts to new contexts, makes decisions under uncertainty, builds trust across difference, and works with systems that are more capable than any previous generation has encountered. They do not replace domain expertise. They make domain expertise renewable.

Four criteria distinguish SuperSkills from ordinary skills. First, durability: a SuperSkill retains value across at least two major technological cycles. Second, transferability: it applies across industries, cultures, and stages of life. Third, AI interaction: it either governs how humans work with intelligent systems or protects against the predictable failure modes automation creates, from judgement decay to skill atrophy. Fourth, compounding effect: it amplifies the effectiveness of other capabilities over time.

Many popular skills fall away under this lens. Creativity without judgement collapses into noise. Technical fluency without ethics scales harm. Resilience without direction becomes endurance theatre. Communication without empathy becomes manipulation. The seven SuperSkills that remain form a coherent system. Each addresses a distinct dimension of human capability that becomes more valuable as machines become more powerful.

The seven SuperSkills#

Curiosity is the disciplined drive to explore, learn, and update beliefs in the face of new evidence. It is not passive openness but active pursuit. In an environment where knowledge expires faster than ever, the disposition to keep learning is foundational rather than optional.

Change Readiness is the capacity to maintain effectiveness while adapting to altered circumstances. It differs from resilience, which emphasises recovery, and from optimism, which emphasises attitude. As transformation becomes continuous rather than episodic, this capacity determines who navigates successfully and who is perpetually destabilised.

Big Picture Thinking is the ability to grasp system interdependencies, long-term patterns, and second-order effects. It enables judgement when local optimisation fails, when immediate actions produce delayed consequences, when the frame that defines a problem determines the quality of solutions.

Empathy is the capacity to understand and respond to others' inner experience while maintaining the distinction between self and other. Sentiment has nothing to do with it. Empathy is the foundation of trust, collaboration and influence. As work becomes more distributed and mediated by technology, the ability to perceive what others think and feel becomes more consequential, not less.

Global Adaptability is the capacity to function effectively across diverse cultural and situational contexts by adjusting approach without losing core identity. As migration, remote collaboration and geopolitical complexity reshape work, the ability to operate beyond one's native context is now a baseline requirement for consequential work.

Principled Innovation is the practice of creating progress under explicit ethical constraint. It rejects the assumption that innovation and responsibility are trade-offs. As the power of new technologies increases, the consequences of unprincipled innovation become more severe.

The Augmented Mindset is the capacity to partner with AI and intelligent tools to extend cognitive capability without surrendering judgement or accountability. It involves knowing when to delegate to machines and when to retain human control, how to evaluate algorithmic outputs, and how to maintain the skills that make human contribution valuable. This is the culminating SuperSkill, because it is where all the others become operational.

Remove any one of the seven, and the system fails in predictable ways. A professional with every SuperSkill except empathy becomes technically effective but relationally corrosive. An organisation with every SuperSkill except principled innovation scales its capabilities and its harms together. A leader with every SuperSkill except big picture thinking optimises brilliantly within a frame that should have been questioned. These seven are the minimum viable set for remaining effective, ethical, and adaptive when intelligent systems handle increasing shares of cognitive work.

Why human distinctiveness increases in value#

A common fear holds that AI advancement diminishes human value. As machines become more capable, humans become less necessary. This fear mistakes the nature of the shift.

What AI advancement diminishes is the value of routine human cognition: tasks that follow predictable patterns, that can be specified algorithmically, that require consistency rather than judgement. What it increases is the value of distinctively human contribution: the judgement that determines whether an output is appropriate for a specific context, the empathy that builds trust in high-stakes relationships, the creativity that generates genuinely novel solutions, the ethics that govern whether a capability should be deployed.

The paradox is straightforward. The more powerful the tools, the more dangerous unskilled human oversight becomes. The more that AI can generate, the more consequential human judgement about what to use becomes. In medicine, diagnostic AI can match or exceed human accuracy on many imaging tasks, but outcomes depend on how clinicians communicate findings and handle the ethics of treatment. In law, generative AI can draft and research at speeds no human can match, but outcomes depend on how lawyers interpret strategic implications and exercise judgement about what matters. In each case, the human contribution becomes more consequential as technological capability increases.

The choice that defines the coming decades#

Here is the implication that runs beneath all of it: in the AI era, capability itself becomes the primary form of inequality.

Those who develop SuperSkills will compound advantage over time. They will navigate change rather than be displaced by it. They will work with powerful tools rather than be diminished by them. They will remain authors of their work rather than executors of algorithmic outputs. Those who do not will find their options narrowing. Not immediately, perhaps. Not dramatically. But steadily, as the gap between the augmented and the dependent widens with each wave of technological advancement.

This work exists to help individuals and organisations move from drift to design. To replace fragile advantage with durable capability. To ensure that as artificial intelligence scales, human intelligence scales with it. The future belongs to those who develop the skills that govern everything else. The time to begin is before the need becomes undeniable.

How the seven work as one#

Seven capabilities held separately are seven things to be good at. Held together they are an operating system, and the argument for why that distinction matters is at how the skills hold together: the defaults of work used to be cultural and are now computational, and authorship erodes through choices nobody made rather than in a single concession.

What the seven produce when they run together is judgement. Each supplies a different component of a decision, and the absence of any one shows up as a recognisable kind of bad decision, which is set out at how the seven skills produce judgement.

The rhythm that keeps them working as one, rather than competing under pressure, is the Five Loops.

About this research#

Written by Rahim Hirji, author of SuperSkills (Kogan Page, 2026), keynote speaker on AI and human capability, and founder of The SuperSkills Intelligence Company. Last reviewed: 7 September 2026.

How this research works · Reviewed quarterly · Found an error? Tell me and it is corrected on the page.

Listen. Heard on Lead with Purpose, Drift or Design: Rahim Hirji On The Question AI Can't Answer for You [015]. Heard on Nothing Ventured, Are We Raising a Generation That Can't Think? | Rahim Hirji. Heard on Solutionary Voices, How to Resist Algorithmic Drift.