Mark Zuckerberg's Meta and Other Hyperscalers Face a Major Bottleneck. Here Are 2 Industrial Stocks That Will Benefit

The market is underestimating how much of the AI spend pool is migrating from semis to infrastructure bottlenecks. That favors CAT and ETN not just on headline orders, but because the scarcity is shifting toward power delivery, on-site generation, switchgear, and project execution capacity — areas where large incumbents can widen pricing and push out smaller rivals. The second-order loser is anyone tied to generic commercial construction or low-value electrical distribution, where AI projects can crowd out less strategic work and absorb labor/material capacity.

The key catalyst window is the next 1-3 quarters, when backlog has to convert into revenue without margin slippage. If CAT/ETN can keep order growth elevated while lead times stay long, estimate revisions should follow; if not, the stocks can de-rate quickly because both are already pricing in a lot of durability. The main near-term risk is that backlog is a lagging signal: project delays, redesigns, or utility interconnection friction would show up before revenue does, and that can matter over days to weeks.

Contrarian take: the consensus is treating AI capex as a clean secular tailwind, but the first derivative may be slowing even if the long-term theme remains intact. CAT looks more exposed to a normalization in earthmoving demand, while ETN is closer to the real bottleneck if grid constraints intensify, so the relative trade is more interesting than an outright long basket. Falsifier: any guide-down in hyperscaler capex, a sharp deceleration in backlog growth, or evidence that AI builds are being delayed by permitting/power availability rather than accelerated by them.