Standard "AI exposure" metrics list automatable tasks but miss a key factor: how tasks relate. If tasks are highly complementary (like steps in cooking), weakness in one part renders the whole output useless. Economists can list tasks but lack data on these crucial interdependencies, limiting the accuracy of job displacement models.

Unlike previous technologies that augmented specific skills, AI could eventually outperform humans in all domains, including creative and emotional tasks. This suggests the historical pattern of technology creating more jobs than it destroys may not hold true.

The fear of mass job replacement by AI is based on a flawed premise. Jobs are not single entities but collections of diverse tasks. AI can automate some tasks but can fully automate very few entire occupations (under 4% in one study), leading to a reshaping of work, not widespread elimination.

The introduction of ATMs unexpectedly doubled the number of bank tellers by enabling banks to open more branches. This historical precedent suggests AI will transform roles in unforeseen ways, shifting tasks from basic functions to relationship-oriented work rather than simply eliminating jobs.

The classic argument that technology always creates new jobs is flawed when applied to AGI. Previous inventions like the tractor automated a single sector. AGI, by its nature, automates all forms of human cognitive labor—from finance to programming—simultaneously, overwhelming society's capacity to retrain and adapt.

Analyzing AI's impact at the job level is misleading. A more nuanced approach is to focus on tasks as the atomic unit of disruption. This allows for a better understanding of how roles will shift and evolve as certain tasks are automated, rather than assuming entire jobs will simply disappear.

Contrary to long-held predictions, AI is disrupting high-status, cognitive professions like law and software engineering before manual labor jobs. This surprising reversal upends the perceived value of higher education and traditional career paths, as the jobs requiring expensive degrees are among the first to be threatened by automation.

The true risk from AI isn't the elimination of titles like "Product Owner," but the automation of repetitive functions. Individuals who merely process tickets or code without understanding business context are becoming obsolete, regardless of their official role.

The first wave of AI job disruption will hit roles that are purely intelligence-based and operate within standardized systems like computers (e.g., software engineering, legal analysis). Jobs requiring physical dexterity in unpredictable, non-standardized environments, like skilled trades, will be automated much later.