AI Attribution Is No Longer a Subtext: Companies Are Saying the Quiet Part Out Loud
TexTak holds [white-collar-displacement] at 70% — up from 67% — and today's news gives us the clearest direct evidence yet for why that move was right. Over 70,000 workers have been displaced by AI in 2026 alone, with PayPal, Coinbase, and General Motors explicitly naming AI automation as the cause. GM's maneuver — laying off 600 IT workers while simultaneously hiring AI-native engineers — isn't a cost-cutting story dressed up in AI language. It's a deliberate skills-swap with a paper trail. That is the forecast target, and it's happening.
Our 70% reflects a specific claim: that companies would move from quiet attrition to public, explicit attribution of workforce reduction to AI. For most of the past two years, the dominant pattern was opacity — headcount declining, AI tooling expanding, no press release connecting the two. The bear case rested on the idea that companies would absorb AI-driven displacement through attrition and natural turnover, never giving journalists or regulators a clean headline. That bear case is now materially weaker.
What makes GM the sharpest data point isn't the number — 600 is not a large layoff in absolute terms. It's the structure: simultaneous hiring of AI-native replacements, explicit framing in internal communications, and no apparent attempt to obscure the mechanism. That's not a company hiding behind 'restructuring.' That's a company treating AI workforce substitution as a boardroom-level strategic communication. When GM does something this explicitly, it signals that the reputational calculus has shifted. The PR risk of attribution is lower than analysts assumed a year ago.
Honestly, the part of our thesis that still keeps us up at night is the 'major layoff wave' framing. Seventy thousand is a large number in aggregate, but it's distributed across dozens of companies. A skeptic can reasonably argue that no single event clears the threshold of a 'wave' with clean attribution — that what we're seeing is more of a distributed reclassification trend than a coordinated moment of institutional acknowledgment. The forecast will resolve cleanly if a single employer announces a layoff exceeding, say, 5,000 roles with explicit AI attribution. We haven't hit that single-event threshold yet, even if the cumulative signal is overwhelming.
What would move us above 80%: a Fortune 100 company — financial services, insurance, or a major bank — announcing a large-scale headcount reduction with AI explicitly named in the earnings call or SEC filing. What would drop us below 55%: a pattern where Q3 earnings calls systematically avoid attribution language despite visible headcount declines, suggesting companies have collectively decided to revert to opacity. We're watching the Q2 earnings cycle closely for exactly that signal.