Citizens reiterates Alphabet stock rating on TPU chip sales outlook By Investing.com

Citizens kept a Market Outperform rating on Alphabet and reiterated a $515 price target, citing TPU chip sales as a new growth driver. The firm now estimates TPU sales of about $3 billion in 2026 and $25 billion in 2027, while lifting 2026 revenue by roughly 1% and 2027 revenue by about 5%. Alphabet shares are already up about 133% over the past year and trade near $383.25, just below the 52-week high of $387.38, suggesting some additional upside if AI monetization accelerates.

Analysis

This is less about a one-time analyst upgrade and more about Alphabet monetizing its infrastructure moat in a way the market has not fully modeled. If TPU sales scale as projected, the important second-order effect is margin mix: revenue tied to proprietary silicon should be far less cyclical than ad spend and much less constrained than external cloud GPU supply, which could re-rate the cloud segment’s quality even if headline growth slows. The key implication is that Alphabet is moving from being a buyer of AI infrastructure to a supplier of it, which can compress customer churn by embedding its stack deeper into enterprise workflows.

The biggest winner beyond GOOGL may be the custom-silicon ecosystem: design tools, advanced packaging, and foundry capacity all gain leverage as hyperscalers diversify away from a single vendor. That is not uniformly bullish for TSM over the medium term; the risk is not an immediate volume hit, but a gradual erosion of pricing power and strategic lock-in if multiple large buyers conclude that custom compute improves unit economics. Over the next 6-18 months, the trade is really about whether AI capex shifts from generic accelerators toward differentiated architectures; if that happens, suppliers with one-customer concentration are the most exposed.

The market may also be underestimating how quickly this can feed back into Alphabet’s own cloud competitiveness. Better internal access to compute lowers latency in product iteration, which can accelerate model launches and improve conversion in enterprise deals; that effect typically shows up with a lag of 2-4 quarters, not immediately in the stock. The main tail risk is political or regulatory: if AI capacity becomes strategically sensitive, governments may push for disclosure, allocation constraints, or export-style oversight that reduces the economic upside of the TPU narrative.

Consensus appears to be treating this as an incremental upside revision, but it may be a broader strategic inflection if TPUs become a monetizable external product. The contrarian view is that the upside could be overdone near term because the market is already pricing in AI leadership, while the real TPU contribution may be back-end loaded and adoption lumpy. Still, if the first external TPU contracts are real and scalable, this is one of the few ways Alphabet can expand earnings power without relying on ad-market acceleration.

Key Decisions for Investors

• Long GOOGL vs short TSM over the next 3-6 months: the pair expresses a view that custom silicon monetization is more additive to Alphabet than it is punitive to diversified compute demand, while TSM faces gradual strategic pressure if hyperscalers keep internalizing chips.
• Add GOOGL on pullbacks over a 2-4 week horizon; upside is driven by multiple expansion if TPU revenue becomes a credible third leg alongside ads and cloud, while downside is limited unless AI capex spend decelerates sharply.
• Buy 6-12 month GOOGL call spreads rather than outright calls: the thesis is a medium-dated rerating as external TPU sales become visible, with defined downside if the market decides the revenue is too small or too delayed.
• For more tactical exposure, short-dated TSM puts or put spreads into any rally tied to AI supply optimism; risk/reward improves if investors start rotating toward in-house accelerator narratives.
• Avoid chasing the move in AAPL on this headline alone; the read-through is indirect and likely needs evidence of supplier diversification cycles before it becomes a tradable catalyst.