AI Automation Creates New Enterprise Risk as Human Expertise Disappears

Quick Facts

• Airbnb CTO warns AI is creating ‘hollowing out’ effect where surface capability remains but human validation expertise disappears
• Entry-level positions that traditionally build expertise are being automated first, breaking the talent pipeline
• Only 24% of enterprises have dedicated AI security governance teams to address these emerging risks

Companies rushing to automate entry-level positions may be destroying the very human expertise they need to validate AI systems, according to Airbnb CTO Ahmad Al-Dahle.

Al-Dahle calls this the ‘hollowing out’ phenomenon, where ‘the surface capability remains (models can still produce outputs that look expert) while the underlying human capacity to validate, extend, or correct that expertise quietly disappears.’

The problem stems from automation targeting junior roles first. ‘Entry-level jobs that develop such expertise were automated first,’ Al-Dahle explains, meaning ‘the next generation of potential experts is not accumulating the kind of judgment that makes a human evaluator worth having in the loop.’

Skills Decay Evidence Mounts

Research supports Al-Dahle’s warnings. Gartner predicts that ‘the atrophy of critical-thinking skills due to over-reliance on generative AI will compel 50% of global organizations to mandate AI-free skills assessments for their employees.’

The World Economic Forum estimates workers can expect ‘39% of their existing skill sets to be transformed or become outdated between 2025 and 2030.’ Skills for jobs have already changed 25% since 2015.

Aviation provides a cautionary precedent. A 2011 FAA analysis found that 60% of accidents involved lack of pilot proficiency in manual operations due to autopilot reliance, a phenomenon called ‘automation-induced skill degradation.’

Four Stages of Decline

Researchers identify a four-stage pattern of cognitive decline. Stage 1 involves experimentation with AI for simple tasks. Stage 2 sees integration into daily workflows. Stage 3 represents reliance where skills begin to atrophy. Stage 4 manifests as addiction where people lose the ability to function effectively without AI assistance.

The resulting ‘cognitive debt’ creates multiple business risks: system fragility when AI fails, quality drift as subtle errors slip through, accountability gaps where leaders approve work they cannot assess, and weak talent pipelines as juniors skip foundational skills.

Al-Dahle warns companies are ‘dismantling the human infrastructure that currently fills the gap, not as a deliberate decision but as a byproduct of a thousand rational ones.’

Governance Gaps Emerge

Only 24% of enterprises have dedicated AI security governance teams, according to recent research. AI introduces ‘unfamiliar risks across multiple disciplines, including compliance, operations, legal, and regulatory’ that are difficult to track as AI deployment becomes decentralized.

Al-Dahle advocates treating ‘the evaluation gap as an open research problem with the same urgency we bring to capability gains.’

The solution is ‘not to make AI painful, but to design AI-empowered processes as developmental rather than frictionless,’ ensuring human expertise continues developing alongside automation.

Read more: The enterprise risk nobody is modeling: AI is replacing the very experts it needs to learn from