Aswath Damodaran argues AI scenarios need probabilities and a human-stakes lens

Aswath Damodaran argues AI scenarios need probabilities and a human-stakes lens

Aswath Damodaran’s post, published March 4, 2026, argues that AI scenarios are more useful when readers distinguish what is possible, plausible and probable, rather than choosing a detailed story or listing outcomes without judging their likelihood. He criticizes the Citrini scenario’s possible rapid, broad worker displacement, while also challenging benign accounts that assume disruption will be limited or slow. Damodaran’s own probable scenario is that AI will reduce workforces in some businesses, bring some game-changing tools, and take longer to unfold—especially in monetization—than advocates expect. He argues that disruption’s gains and costs depend on its form and speed: replacing workers may raise productivity but also remove income and spending power, while slower change gives workers more time to adjust. He warns that benefits may concentrate among a few businesses and individuals, worsening inequality. He also stresses that work brings self-esteem and a sense of worth, so displacement can cause “psychic damage,” and argues that society may respond with schadenfreude because earlier displaced workers were largely ignored. His closing advice is to prepare for an AI imitator by building skills and qualities that make one necessary, even if disruption fizzles. How likely is his probable scenario, given the uncertainties he describes?

Publication: 7a843979-59ee-c73e-cf6d-17696fa277c7

attribution — pass: The reading is a single blog post by Aswath Damodaran and is attributed to him throughout. The views of Citrini, Citadel and others are kept separate from his own.

my intent in this post is less about promoting my favored AI scenario, and more about providing a framework for you to develop your own

summary — uncertain: Two passages need repair. First, the summary never says who Citrini is: a macro and stock research publisher whose report imagined unemployment above 10% and a market drop of almost 40% by June 30, 2028. A general reader cannot weigh the critique without that. Second, it says he challenges benign accounts that assume disruption will be 'limited or slow', yet he himself expects slower unfolding. He rejects the marginal-tools fizzle and the no-displacement view, not slowness. Smallest repair: identify Citrini briefly and replace 'limited or slow' with 'that AI will only marginally help workers or not displace jobs'.

it will take longer to unfold, at least when it comes to monetization, than its advocates think.

date — pass: The instant 2026-03-04T17:10-05:00 is March 4, 2026 in UTC, matching the stated date at day precision. Nothing is presented after that date, and Citrini's 2028 figures are reported as a scenario, not an outcome. The page was modified after publication (Oct 2026). However, the cited passages are core body argument consistent with the March 2026 context.

The Citrini AI assessment came out on February 22, 2026, and it starts with a preface stating that it is presenting a scenario, not a prediction.

support — pass: Each claim matches a verbatim excerpt and is framed as Damodaran's argument or warning. The inequality, self-esteem and Pascal-style preparation points are source-stated. The limitations correctly note his self-described limited AI and macro knowledge and the missing table and graph. The cited passages are plainly part of the original argument despite the later page modification.

it is possible that the net benefits accrue to a few businesses (and individuals), leaving the rest (businesses and individuals) with net costs.

media — pass: No media is selected, and no verified media candidates were supplied, so there is no identity or relevance issue.