How accurate have Ed Zitron’s AI skeptic predictions been?

The repeated failure of prominent AI skeptic Ed Zitron’s predictions against actual financial and operational data raises serious questions about the analytical rigor behind one of the most circulated anti-AI narratives in the technology press. A systematic review of Zitron’s forecasts, conducted by an independent analyst with no financial stake in AI outcomes, reveals a consistent pattern of dire warnings that have not materialized, supported by reasoning that misreads the underlying health of the companies he critiques.

Consider Zitron’s November 2024 claim that major technology companies like Meta and Google are dying and thrashing around on artificial intelligence because they have run out of growth options. He specifically called Meta a dying product and a dying company, and argued that none of these companies know how to grow anymore, suggesting their embrace of AI is a desperation move to reignite growth in a dying ecosystem. The financial data tells a different story. Meta’s revenue and GAAP operating income have been on an upward trajectory, with no signs of the existential decline Zitron forecasts. Alphabet and Microsoft have posted similarly strong revenue and profit numbers. These are not the figures of dying companies grasping at straws. They are the results of enterprises that are growing rapidly across multiple business lines, with AI representing one of several strategic investments rather than a Hail Mary pass.

The pattern repeats across Zitron’s broader body of predictions. He identifies minor issues-such as a purported drop in Facebook’s monthly active users-and extrapolates them into catastrophic conclusions that the data does not support. This is not the mark of a careful skeptic but of a commentator who has adopted a conclusion and works backward to justify it. Skepticism toward AI is intellectually legitimate and often healthy. The technology has real limitations, genuine risks, and legitimate critics whose concerns deserve attention. But skepticism that consistently contradicts observable reality, that treats healthy corporate growth as evidence of impending death, ceases to be skepticism and becomes something closer to reflex.

The deeper problem is that this kind of analysis does not correct itself. When predictions fail, the framework remains intact, ready to generate the next set of dire forecasts that will also likely miss the mark. Investors and policymakers who rely on such commentary risk making decisions based on a worldview that is systematically disconnected from the data. The AI debate is important enough that it deserves critics who are wrong for the right reasons, not critics whose errors are baked into the structure of their reasoning.

The takeaway for professional readers is straightforward. Treat all AI forecasting, whether boosterish or skeptical, with the same evidentiary standard. Demand specific, falsifiable predictions and hold forecasters accountable when those predictions fail. A healthy market for ideas depends on it.