In the interest of providing multiple means of engaging with this blog, below is the podcast style conversation, using our AI friends Johnny and Joanne who go into each blog post at a deeper level. Or below is the usual written format which is a little more TL;DR
With my colleagues, we have recently been working on a challenge to give groups during the upcoming leadership development week. This understandably was focused around the most profound challenge of our time - and the question that lives rent-free in mine and many colleagues’ minds- how does generative AI differ from previous automation? This is fundamentally different from every previous automation wave, and understanding why requires us to think deeply about the nature of human value itself. This is no dystopia, but we do need to think about how we organise work around this different form of automation.
Let me walk you through this step by step, because grasping this shift is crucial for anyone trying to navigate the AI transformation.
For decades, we've had a comfortable understanding of which jobs were safe from automation. Like a well-worn map, we could plot the territories where humans would always have the advantage. Physical jobs requiring dexterity and judgment? Safer. Creative work requiring innovation and human connection? Definitely safer. Complex mental tasks involving unique circumstances and empathy? Absolutely safe.
That map has just become obsolete.
The arrival of generative AI hasn't simply automated another category of work—it has fundamentally rewritten the rules of what can be automated. To understand why this matters so profoundly for how we think about work and workforce planning, we need to examine how this technology has shattered the traditional boundaries that protected human work from automation.
The Three-Dimensional Fortress of Human Work
Since the Industrial Revolution, automation has followed predictable patterns. We could map any job along three key dimensions that historically determined whether it was vulnerable to automation or remained firmly in human territory.
The Repetitive-Variable Dimension. Traditional automation excelled at repetitive tasks with predictable routines and clear success criteria. Think of factory assembly lines or data entry. Meanwhile, variable work—tasks requiring innovation, judgment, and the application of decision rules to unique circumstances remained human territory. A surgeon performing a routine procedure might use automated tools, but adapting to unexpected complications during surgery requires human judgment.
The Independent-Interactive Dimension Machines could handle independent tasks performed by a single person with minimal communication requirements. But interactive work involving collaboration, communication, and empathy remained distinctly human. Customer service representatives, team leaders, and counsellors were protected by this interactive barrier.
The Physical-Mental Dimension Perhaps most obviously, mental work requiring cognition, creativity, and judgment was considered the ultimate human refuge as machines took over physical labour requiring strength and manual dexterity. The progression seemed clear: humans would move from physical work to mental work as automation advanced (Jesuthasan & Boudreau, 2018).
These three dimensions created what felt like a fortress around human cognitive work. Jobs that scored high on the variable, interactive, and mental dimensions were considered automation-proof. This framework shaped decades of workforce planning, career advice, and educational policy.
Understanding the Historical Escalator
Think of the history of automation as an escalator that has been carrying humans upward through different types of work. For centuries, this escalator moved predictably: when machines took over physical tasks, humans moved to mental tasks. When computers automated routine mental tasks, humans moved to creative and strategic mental work.
The pattern seemed reliable. Tractors replaced farm labourers, so people became factory workers. Assembly lines replaced factory workers, so people became office workers. Computers replaced clerks, so people became analysts and managers. Each time, humans climbed higher up what we might call the "cognitive value chain."
But here's what makes generative AI different: it's not just automating the bottom rungs of mental work - it's placing a ceiling somewhere in the middle of what we thought was the human domain. This forces us to reimagine where human value lies completely.
The New Frontier: Beyond Intelligence to Wisdom
To understand where humans move next, we need to distinguish between intelligence and wisdom. Intelligence is about processing information, recognising patterns, and applying learned knowledge. Wisdom is about understanding the deeper meaning, navigating complexity with good judgment, and making decisions that account for long-term consequences and human values.
Generative AI excels at intelligence tasks. It can process vast amounts of information, identify patterns, and even engage in sophisticated reasoning. But wisdom requires something more - it requires the integration of knowledge with lived experience, values, and an understanding of human nature that comes from being human.
The Five Domains of Irreplaceable Human Value
Let me outline where I see humans moving in this new landscape, building from the most concrete to the most abstract:
First, Complex Contextual Judgment. While AI can analyse data and even make recommendations, humans excel at understanding the full context that data cannot capture. Consider a doctor using AI diagnostic tools. The AI might identify potential conditions based on symptoms and test results, but the human doctor integrates this with their understanding of the patient's life circumstances, family history, psychological state, and cultural background to make treatment decisions that truly serve the whole person.
For your organisation, this might mean engineers who use AI for technical calculations but apply human judgment to understand how a solution will work in the real-world political, environmental, and operational context of a specific client's situation.
Second, Values-Based Decision Making. AI can optimise for specific metrics, but it struggles with situations that require balancing competing values or making ethical judgments in ambiguous situations. Humans excel at navigating the grey areas where there's no clear "right" answer, only better or worse approaches given the values and priorities at stake.
Think about urban planning. AI can optimise traffic flow or energy efficiency, but human planners must balance efficiency with equity, economic development with environmental protection, and individual convenience with community cohesion. These decisions require understanding not just what is technically possible, but what is morally and socially desirable.
Third, Deep Relationship Architecture. While AI can engage in conversations and even demonstrate empathy, humans excel at building the kind of deep, trust-based relationships that enable complex collaboration and long-term partnerships. This goes beyond customer service to what we might call "relationship architecture" - designing and nurturing the human connections that make complex organisations and societies function.
For your organisation, this might mean relationship managers who don't just sell products but truly understand their clients' strategic challenges and can architect long-term partnerships that evolve over time. They use their human understanding of trust, politics, and organisational dynamics to create value that goes far beyond any individual transaction.
Fourth, Creative Synthesis and Innovation. Here's where we need to be nuanced. AI can certainly be creative in combining existing ideas in novel ways. But humans excel at the kind of breakthrough thinking that comes from integrating insights across completely different domains, questioning fundamental assumptions, or imagining entirely new possibilities.
Human creativity often comes from the intersection of professional expertise with personal experiences, cultural background, and unique perspective. An engineer might solve a complex technical problem by drawing inspiration from their hobby as a musician, their experience raising children, or their cultural background - connections that AI, despite its vast training, cannot make because it lacks lived experience.
Fifth, Systems Leadership and Orchestration. Perhaps most importantly, humans are moving toward what I call "systems leadership" - the ability to orchestrate complex ecosystems of people, processes, and technologies to achieve outcomes that none could achieve alone.
This isn't traditional management or even strategic planning. It's the ability to see patterns across complex systems, understand how different stakeholders think and what motivates them, and design interactions that align diverse interests toward common goals. It's the uniquely human capacity to inspire, influence, and orchestrate without direct control.
The Meta-Skill: Teaching Machines to Serve Human Purposes
Interestingly, one of the most valuable human skills emerging is the ability to effectively collaborate with AI systems themselves. This isn't just about prompt engineering or using AI tools - it's about understanding how to combine human insight with AI capabilities to achieve outcomes neither could reach alone.
Think of this as a meta-skill: humans who can teach machines not just what to do, but how to serve human purposes in ways that align with human values and wisdom. These humans become the interface between artificial intelligence and human flourishing.
Practical Implications for Job Design
I suggest several key areas where new roles and value might emerge:
Wisdom Integrators - Professionals who specialise in taking AI-generated analysis and integrating it with contextual understanding, stakeholder perspectives, and long-term strategic thinking to make decisions that serve broader human purposes.
Relationship Architects - People who focus not just on individual client relationships but on designing and nurturing entire ecosystems of partnerships, collaborations, and stakeholder networks that create sustained competitive advantage.
Values Navigators - Specialists who help organisations navigate the ethical and societal implications of their technology decisions, ensuring that efficiency gains don't come at the cost of human values or social responsibility.
Innovation Catalysts - Individuals who excel at spotting opportunities for breakthrough innovation by connecting insights across different domains, industries, and ways of thinking.
System Orchestrators - Leaders who can manage complex projects and initiatives that involve multiple organisations, technologies, and stakeholder groups, using their understanding of human psychology and organisational dynamics to achieve alignment and results.
The Encouraging Reality
Here's what I really want you to understand: while generative AI has indeed moved the boundary of what machines can do, it has also made human wisdom, judgment, and relationship-building skills more valuable than ever. The companies that will thrive are those that figure out how to combine AI's information processing power with uniquely human capabilities.
The key insight is that we're not running out of valuable work for humans - we're being forced to discover what human work is most uniquely valuable. This can be unsettling, but it's also an incredible opportunity for those who embrace the challenge of moving to this higher level of human contribution.
Jesuthasan, R., & Boudreau, J. (2018). Reinventing Jobs: A 4 Step Approach for Applying Automaton to Work. HBR Press.