AI Automation and the Fallacy of Job Exposure Scores
An analysis of why numeric AI job exposure scores are misleading and how to identify true value drivers in business.
The Illusion of Precision in AI Automation
Many current frameworks used to predict which jobs will be displaced by AI are fundamentally flawed. The tendency to assign numeric 'exposure' scores to specific roles—such as claiming an accountant has a 72% exposure rate—is a form of self-deception. This precision is misleading because it fails to account for the complexity of what a job actually entails beyond simple task automation.
Identifying the 'Job to be Done'
To understand the impact of technology, one must look past the linear extrapolation of current tasks. The 'Uber test' suggests that while some technologies simply make existing processes better, others redefine the entire market. For example, while the internet automated booking, it didn't replace airlines because the core product—transporting a person from A to B—remained physical. In contrast, retail and journalism were disrupted because their core value proposition was often tied to the physical distribution of goods or information, which the internet could bypass.
The Enduring Value of the Human Element
History shows that automation often increases the demand for certain professions. Accountants, despite wave after wave of automation (from adding machines to Excel), have seen their numbers grow throughout the 20th century. This is because the 'job to be done' for an accountant is not merely adding numbers, but providing professional judgment and trust.
In the AI era, the competitive advantage shifts from the technical ability to write code or generate documents to the human elements of consultancy, curation, taste, and strategic implementation. The ability to map complex technology to specific business needs will be the primary differentiator for companies and professionals alike.
Key insights
-
Numeric scores for AI job exposure are misleading because they are based on a simplification of what a job is. They mistake the tasks (the 'how') for the value proposition (the 'why').
Impact: Businesses may misallocate resources or fail to plan for real disruption by focusing on task-based automation rather than value-chain transformation.
-
The 'Uber Test' highlights that true disruption occurs when technology changes the total addressable market (TAM) rather than just improving a current process. Linear extrapolation from existing markets is often wrong.
Impact: Entrepreneurs can identify new market opportunities by asking how technology fundamentally changes the customer's 'job to be done' rather than just automating a current task.
-
Automation typically impacts the delivery method rather than the core value. For instance, the software development process is easier, but the 'hard part' remains the product-market fit, sales, and strategic execution.
Impact: Software companies may find their technical barriers to entry lower, reducing the product's uniqueness based on code alone and increasing the importance of go-to-market strategy.
-
Physicality is not the primary determinant of safety from automation; rather, it is whether the physical aspect of the product is the actual value driver for the customer.
Impact: Industries relying on physical assets as a barrier to entry may be more vulnerable than those where physicality is the core product.
-
As technical capabilities become unfathomable and ubiquitous, the value shifts toward the 'white glove' experience, consultancy, and human-led implementation.
Impact: High-end consultancy and professional services may see a premium on 'taste' and expert human judgment over the raw output of AI tools.
Action items
-
Conduct a 'Job to be Done' analysis of current business operations to separate the delivery mechanism (the 'how') from the value proposition (the value customers pay for).
Impact: Allows companies to identify which parts of of their business are truly vulnerable to AI automation and which parts provide enduring value.
-
Shift focus from technical feature parity to human-led implementation and strategic consultancy services to maintain competitive advantage.
Impact: Increases customer retention and high-ticket deal closure by focusing on the human element of solving complex organizational problems.
-
Avoid relying on numeric 'AI exposure' benchmarks for strategic planning; instead, use qualitative frameworks that analyze the point of leverage in the value chain.
Impact: Prevents strategic errors based on false precision and encourages deeper analysis of industry-specific market dynamics.
Quotes
“Trying to put a numeric score on it, it is idiotic. You know, people now trying to say, well, know, accountants have got exposure of 72 and bookkeepers have got an exposure of 84. This is just insane.”
“The question isn't physicality, is your product physical or not? The question is, does physicality matter?”
“The map that's not the maths, yeah. That's not what the job is.”