Big tech faces AI spending scrutiny ahead of Q2 earnings: Wedbush

Investors are heading into Q2 earnings with the key question of when record AI infrastructure spending by major tech companies will start producing stronger revenue growth. Wedbush said recent weakness in large-cap technology stocks is driven by timing concerns over returns, not a worsening long-term outlook. The article is largely a sentiment check on AI capex and earnings expectations rather than a new company-specific development.

Analysis

The market is effectively pricing a timing problem, not an impairment problem: the key question this earnings season is whether AI capex starts showing up in operating leverage quickly enough to justify current multiples. That creates a bifurcation between compute enablers and the platform names funding the buildout — the former can continue to see order visibility even if monetization lags, while the latter face multiple compression if investors conclude returns are 2-4 quarters further out than expected.

Second-order beneficiaries are likely to be the picks-and-shovels layer with the shortest revenue conversion cycle: networking, power infrastructure, thermal management, and memory/advanced packaging vendors should remain relatively insulated because their demand is tied to build schedules, not end-user AI adoption. The vulnerable cohort is any mega-cap that must keep spending aggressively while showing only incremental revenue acceleration; those names are at risk of becoming “capex traps” where each quarter of disappointment forces higher hurdle rates and lower terminal-margin assumptions.

The near-term catalyst window is the next 1-2 earnings prints: if commentary shifts from “build first” to “optimize and monetize,” the trade can rip higher quickly because positioning is crowded and underexposed to upside surprises. Conversely, a mixed report with strong spend but weak usage conversion could trigger another leg down in large-cap tech over days to weeks, especially if management teams avoid giving explicit payback timelines. Over a 6-12 month horizon, the setup improves if AI inference demand broadens beyond hyperscalers into enterprise software and vertical workflows; if that does not happen, the market will increasingly treat AI infrastructure like a cyclical capacity boom rather than a durable growth driver.

The contrarian read is that the selloff in large-cap tech may already reflect a lot of this skepticism, while the real fragility sits in the supply chain names that have assumed an uninterrupted ramp. Consensus is focused on whether AI spend pays off; the sharper question is whether the market has underpriced the probability of a temporary digestion phase where beneficiaries keep ordering but customers pause on new commitments. That would favor a rotation from the most extended infrastructure exposures into quality platform names with the strongest balance sheets and the clearest path to monetization.

Key Decisions for Investors

• Stay long the AI infrastructure supply chain versus large-cap platform tech for the next 4-8 weeks: prefer names with book-to-bill visibility and short revenue recognition lag; avoid the highest-multiple hyperscalers until earnings confirm monetization inflection.
• Use put spreads on the most crowded mega-cap AI spenders into earnings as a tactical hedge: 1-3 month tenor, defined risk, looking for a 10-20% downside move if guidance emphasizes continued capex without clear revenue uplift.
• Pair trade: long semiconductor / networking / power-electrification beneficiaries, short an equal-dollar basket of mega-cap tech with the heaviest AI capex intensity; thesis is that the supply chain can compound on backlog while the funders face multiple compression.
• If earnings show better-than-expected AI revenue traction, rotate quickly into the highest-quality platform names on a 6-12 month horizon; upside could be 15-25% as the market de-risks payback timing and expands terminal assumptions.
• Watch for any management language about slowing incremental capex growth — if that appears, take profits on the supply-chain trade immediately, because the market will likely re-rate the entire AI complex lower on fears of demand normalization.