2 AI Stocks Jim Cramer Says Are Still Worth Buying After Massive Gains

Jim Cramer remains bullish on Nvidia and Micron after five-year gains exceeding 1,030% and 1,410%, respectively, citing Nvidia’s AI-chip position and Micron’s supply-constrained memory demand. Nvidia raised its remaining buyback authorization by $150 billion to $235 billion, while Micron beat analyst estimates and issued above-consensus current-quarter guidance. The article cautions that Micron’s memory business is cyclical and that both stocks depend on continued AI infrastructure spending.

Analysis

The key distinction is durability of rents: NVIDIA’s software ecosystem may help preserve economics across chip cycles, while Micron’s returns depend more directly on a supply-constrained memory market staying tight. That makes Micron’s low forward multiple a poor standalone value signal: if pricing and utilization are near a cyclical peak, estimates can fall faster than the multiple looks cheap. Conversely, rapid HBM demand does not guarantee industry-wide memory scarcity; capacity decisions can eventually loosen conventional DRAM and NAND markets as well.

NVIDIA’s buyback authorization is not equivalent to executed repurchases and cannot offset a deterioration in customer returns on AI infrastructure. The near-term test is whether major cloud customers continue to fund data-center buildouts and demonstrate monetization, not the authorization headline. Over 1–3 months, earnings guidance and hyperscaler capex commentary matter more than pundit conviction. Over 6–18 months, custom accelerators, inference-efficiency gains, and memory capacity additions could reduce the intensity of demand per unit of AI workload.

Contrarian view: the market may be underpricing how long AI infrastructure investment persists, but it may also be extrapolating current memory scarcity too far. Prefer NVIDIA’s relatively more durable competitive position, while treating Micron as a higher-beta, cycle-sensitive expression. No price targets are warranted from the supplied information; verify current valuation, customer capex plans, and Micron’s HBM mix and supply outlook before sizing.

Key Decisions for Investors

• Prefer NVIDIA to Micron for incremental AI exposure; consider building NVIDIA on weakness rather than chasing strength. Reassess if data-center revenue growth or gross-margin trends materially weaken, or if major cloud customers cut capex guidance.
• Avoid adding aggressively to Micron solely on its forward P/E. Treat it as a tactical position and scale in only if upcoming results confirm sustained HBM demand without a broadening memory supply response; monitor inventory, pricing, and company capacity plans.
• A relative-value long-NVIDIA/short-Micron trade is a possible expression of software-moat durability versus memory cyclicality, but keep it modest: both remain exposed to a shared AI-capex reversal. Revisit if Micron’s supply discipline and HBM profitability prove durable while NVIDIA’s customer spending indicators deteriorate.
• Watch for falsification over the next 1–3 months: hyperscaler capex reductions, weaker AI infrastructure guidance, or evidence of memory inventory accumulation. These would challenge the demand thesis; sustained spending and tight memory pricing would argue against fading either name.