The Silence Is the Signal: AI Displacement Is Real, But the Forecast Clock Starts When a CEO Says So

Let's be precise about what we're predicting and why prior waves don't satisfy it. The 2023 tech layoff wave — 200,000+ workers in Q1 alone — included companies citing 'efficiency' and 'right-sizing' in earnings calls, with some referencing AI tooling in passing. We reviewed that evidence carefully when setting this forecast. It does not meet our bar. 'AI as part of a broader efficiency narrative' is categorically different from 'we reduced headcount because AI now performs these functions.' The 2023 wave was primarily over-hiring correction from pandemic-era demand surge. The 2024 wave was continuation of the same correction cycle. What we're forecasting is the first time a major employer uses SEC-filed language or earnings call testimony to say, in substance: these roles were eliminated because AI performs them adequately. That has not happened yet, which is why the forecast remains open.

So what does today's evidence actually prove? The 150,000 displacement figure and the specific cuts at Amazon (16,000 corporate roles) and Meta (8,000 planned) are proximate evidence — they tell us the scale of displacement is large enough that explicit attribution becomes statistically more likely. Think of it this way: if 10 companies are quietly replacing workers with AI, the probability that at least one executive eventually says so in a formal filing is higher than if only two are. The Yale data on entry-level compression — recent graduates aged 22-27 running 5.6% unemployment against 4.2% overall — is circumstantial but points to something structurally different from prior cycles. Companies are not just laying off; they are not hiring in the first place. That behavioral pattern is consistent with AI substitution at the entry level. It is not proof of it.

Why 70% and not 55% or 85%? We're weighting three things heavily. First, the scale of the current wave. When displacement reaches 150,000 workers in four months while AI capex simultaneously hits $725B, the gap between what is happening operationally and what is being said publicly becomes increasingly hard to maintain. Investor relations teams face direct questions about AI ROI. The pressure to demonstrate that the capex is generating headcount savings — not just productivity gains — will eventually pull explicit attribution into formal filings. Second, competitive pressure between firms. If one major employer claims AI productivity as a structural advantage in an earnings call, peers face pressure to match the narrative. Attribution cascades once it starts. Third, the timeline: we have not set a hard deadline in this piece because the forecast on TexTak carries its own resolution window, but we're watching Q2 and Q3 2026 earnings cycles as the most likely trigger moments. The 70% does not yet account for legal/HR liability suppression — which is the most powerful force working against resolution. Lawyers routinely advise executives not to attribute cuts to automation because it creates wrongful termination exposure. That headwind is real and partially why we're at 70% rather than 80%.

The part that keeps us up at night is Sam Altman's AI washing observation — and it cuts both ways against our forecast. Altman acknowledged companies are blaming cuts on AI that would happen anyway. If that's true at scale, two problems emerge: first, explicit attribution may arrive not because AI caused displacement but because executives find it a convenient narrative for cost-cuts they'd make regardless. That would resolve our forecast technically while rendering it analytically meaningless. Second, the February NBER study finding that 90% of C-suite executives report no AI employment impact over three years is not easily dismissed as executives lying. If we take it at face value as a base-rate anchor, it suggests that even now, with models deployed across enterprise workflows, the causal connection between AI and headcount reduction is genuinely ambiguous to the people making the decisions. Our 70% requires a mechanism that produces clear public attribution despite that ambiguity. The most plausible mechanism: a major employer in a non-tech sector — financial services, logistics, insurance — where AI substitution is less defensible as 'efficiency' and more nakedly structural, makes a filing that becomes a landmark. That specific pathway is what we're watching. If Q2 and Q3 earnings cycles pass without it, we would revisit downward toward 60%.