AI Infrastructure: Winners and Losers in Data Center Buildout

Barclays Upgrades Vertiv to Overweight, Raises Price Target to $200What happenedBarclays upgraded $VRT (Vertiv) from Equal Weight to Overweight and lifted its price target to $200 (from $181), saying recent volatility created an attractive entry point and that $VRT could “catch up” versus other AI-infrastructure names into 2026.Why it matters for tradersThis is another “AI data center build-out is still real” signal. If investors rotate further into “picks-and-shovels” (power, cooling, electrical gear), it can spill over to the whole data-center infrastructure complex.WinnersData center power + cooling “picks-and-shovels”Analyst upgrade momentum + continued AI rack density trends mean more spend on thermal management, power distribution, and liquid cooling where $VRT is heavily exposed.Names: $VRT (Vertiv Holdings Co), $ETN (Eaton Corp), $NVT (nVent Electric plc), $JCI (Johnson Controls International plc)Backup power, generators, and electrical equipment leveraged to data center build-outsHigher AI/data-center power needs drive demand for generators, switchgear, and electrical components to keep facilities online and scalable.Names: $CAT (Caterpillar Inc), $GNRC (Generac Holdings Inc), $HUBB (Hubbell Inc), $EMR (Emerson Electric Co)Data center landlords and colocation operatorsIf “capex slowdown” fears fade, the underlying takeaway is sustained demand for data center capacity—supportive for operators that lease space and power to customers.Names: $EQIX (Equinix Inc), $DLR (Digital Realty Trust Inc), $IRM (International Business Machines Corp)LosersHyperscalers facing heavier capex and power constraintsMore AI data center build-out can mean higher capex intensity, higher power bills, and potential margin pressure—especially as power availability becomes a bottleneck.Names: $AMZN (Amazon Inc), $MSFT (Microsoft Corp), $GOOGL (Alphabet Inc)Legacy enterprise IT spend that can get displaced by cloud/AI infrastructure cycles (relative underperformance risk)As budgets shift toward AI infrastructure and cloud capacity, some “traditional enterprise refresh” areas can lag in relative performance. (Market-rotation thesis, not a direct call on fundamentals.)Names: $CSCO (Cisco Systems Inc), $HPE (Hewlett Packard Enterprise Co), $HPQ (HP Inc)Utilities with grid-upgrade burden and execution risk in heavy data center regionsAI-scale power demand (and long interconnection timelines) can force large grid capex and complex rate cases; some utilities may see headline risk around costs and timelines.Names: $ED (Consolidated Edison Inc), $PCG (PG&E Corp), $EIX (Edison International)#StockMarket #Trading #Investing #DayTrading #SwingTrading #AI #DataCenters #Infrastructure #PowerManagement #Cooling #ElectricalEquipment