3 AI Growth Giants That Are Looking Like Value Stocks

The article argues that Nvidia, Alphabet, and Micron offer AI-driven growth at value-like valuations after recent share-price declines. Nvidia is cited at 22x forward earnings after revenue and net income rose to $96B and $59B, while Alphabet trades at 16x forward earnings alongside reported 82% Google Cloud revenue growth to more than $24B and 1B Gemini monthly active users. Micron is presented as the deepest value candidate at 6x forward earnings, supported by more than 300% quarterly revenue growth to $41B and memory-supply constraints.

Analysis

The apparent valuation convergence is not equivalent across the three names. NVDA’s multiple is primarily a duration debate: the market is discounting a lower terminal growth rate and eventual custom-ASIC substitution, not a near-term demand collapse. GOOG’s discount instead reflects uncertainty over whether AI raises search monetization and cloud margins or cannibalizes query economics; that makes it the cleaner re-rating candidate if Search revenue-per-query and Cloud operating margin hold simultaneously. MU’s low multiple is least informative because memory earnings are cyclical and consensus estimates are unusually sensitive to HBM pricing, yields, and the next supply-addition cycle.

The second-order beneficiary of sustained accelerator demand is not necessarily another GPU vendor but the memory-content stack: MU and SK Hynix exposure should gain as inference shifts from training clusters toward memory-intensive agentic workloads. The offset is that a successful broad deployment of Google TPUs or hyperscaler ASICs could reduce NVDA unit share while expanding total HBM demand, creating a relative long MU/short NVDA hedge. Conversely, a capex digestion phase would hurt MU disproportionately: memory pricing typically turns before end-demand visibly weakens, whereas NVDA retains a larger software, networking, and system-level profit pool.

The article’s operating figures should not be used as an underwriting input without primary-source reconciliation; several cited magnitudes appear inconsistent with normal quarterly reporting scale. Near term, this is largely retail-friendly valuation commentary rather than a new fundamental catalyst. Over 1-3 months, the relevant catalysts are hyperscaler capex guidance, NVDA supply/lead-time commentary, Google Search monetization disclosures, and HBM contract-pricing evidence; over 6-18 months, the key question is whether AI revenue becomes externally monetized rather than remaining an internal infrastructure spend.

Contrarian view: “cheap AI” may be a misleading frame where forward earnings embed peak pricing or unusually favorable mix. GOOG is the strongest value-with-optionality expression because its core cash flows can support the multiple even if AI returns disappoint; MU requires confidence that the memory upcycle persists through new capacity. A broad AI selloff driven by rates or capex skepticism would likely compress NVDA first on sentiment, but earnings-estimate risk would ultimately be greatest in MU.

Key Decisions for Investors

• Prefer long GOOG over NVDA on a 6-12 month horizon: accumulate on AI-led weakness while monitoring Search revenue growth and Google Cloud operating margin. Thesis is falsified by two consecutive quarters of decelerating Search monetization alongside margin erosion; expected payoff is multiple normalization plus Cloud upside, with lower cyclicality than semiconductor exposure.
• Use a 3-6 month pair trade: long MU / short NVDA in equal beta-adjusted dollars only if HBM pricing and customer qualification data remain firm. This expresses rising memory content and ASIC substitution; stop if MU guides materially lower gross margin or NVDA reports sustained accelerator lead-time expansion and upward revenue revisions.
• Do not buy MU solely on headline forward P/E. Establish an alert around DRAM/NAND spot-price momentum, HBM yield commentary, and announced capacity additions from Samsung and SK Hynix; a turn in pricing or accelerated supply additions would warrant avoiding or reducing exposure before consensus EPS resets.
• For NVDA holders, retain core exposure but fund downside protection through 3-6 month put spreads around the next hyperscaler capex-reporting window. The central risk is not an immediate revenue miss but a reduction in 2027-28 spending expectations, which could drive multiple compression before reported earnings weaken.