Contents · 5 sections+
The Bureau of Labor Statistics and Gemini Flash recently collaborated on something worth sitting with. They scored 143 million US jobs against AI exposure, one to ten, and the result is not the headline most people are writing.
The average came out at 4.9. A near-perfect midpoint. Which would be reassuring, if averages told you anything useful about the tails.
They don't.
I.The money is in the tails.
Roughly $3.7 trillion in annual wages sits in jobs scored seven or above. That is not a rounding error. That is the productive core of the knowledge economy, the functions that generate, manage, and protect enterprise value. And 42 percent of the US workforce falls into the high or very high exposure categories, against only 33 percent with low exposure.
The comfortable middle is smaller than the narrative suggests.
II.Education cuts the wrong way.
Here is where the data stops being kind to received wisdom. Workers with bachelor's degrees carry an exposure score of 6.7. Workers without a degree or with only a high school diploma sit at 4.1.
The credential economy built its premium on tasks that AI can now perform at scale: synthesis, summarisation, structured analysis, pattern recognition across large data sets. The degree did not confer irreplaceability. It conferred access to exactly the work that is now being automated.
This matters for executive search because the pipeline of talent that organisations have trusted for decades was educated and evaluated against a set of competencies that is being repriced in real time.
III.Pay is a proxy for exposure.
Jobs earning above $100,000 carry an exposure score of 6.7. Jobs earning below $35,000 score 3.4.
The implication is direct. The executives your organisation needs most, the ones drawing the largest compensation packages, are operating in the highest-exposure functions. Not because they are at risk of replacement in the near term, but because the nature of what makes them valuable is shifting underneath them.
The question for a CEO or board is not whether AI will touch their leadership team. It already has. The question is whether the people currently in those seats have reconfigured how they work, or whether they are still producing the same outputs more slowly than a well-prompted model.
IV.A case worth naming.
The data flags wholesale and manufacturing sales representatives with a score of seven. The reasoning: AI now handles lead generation, personalised outreach, and price optimisation with sufficient competence to reduce headcount requirements. Physical trade shows and negotiation still offer some protection, for now.
The same logic applies, one level up, to every revenue leadership role that has not yet confronted what the function actually requires when the first three layers of the sales motion are automated. The title does not change. The job description does.
V.What this means for the mandates we take.
At Sercxi, we run every assignment through the APEX Intelligence Engine before a single approach is made. Part of what that scores is entropy, the degree to which a function is structurally disrupted relative to the capability of the incumbent.
The exposure data confirms what we have been observing operationally across data centre, AI infrastructure, and FinTech leadership in APAC and EMEA: organisations are not short of executives. They are short of executives who have adapted their decision-making to the environment they are actually in.
That is a different brief. And it requires a different kind of search.
The Sercxi Displacement Index tracks AI-driven role disruption across the sectors we cover. If you are assessing a leadership team or planning a hire in a high-exposure function, the brief starts here.