My Top 3 Predictions for AI Stocks in the Second Half

AI-related stocks are regaining momentum as valuations cool and AI demand remains strong, with the S&P 500 gaining nearly 15% in Q2 (best quarterly performance in six years). The article’s outlook is that Nvidia could see a “burst of growth” in H2 as it launches the Vera Rubin GPU platform and targets the ~$200B CPU market, including nearly $20B in standalone CPU sales guidance for this year. For SpaceX, IPO-driven enthusiasm may fade as heavy investment continues (last year’s loss was $4.9B), while investors are expected to broaden exposure across more stocks to manage AI-related risk.

Analysis

The market is shifting from paying up for any AI exposure to paying for proven monetization plus platform optionality. That favors the category leader with real switching costs, because even a modest deceleration can still look like growth scarcity relative to the rest of tech; the risk is only if hyperscaler capex starts to roll over, which would hit multiple before it hits the P&L.

A wider AI rally also changes the plumbing: if participation broadens, liquidity franchises should quietly benefit from higher turnover and derivatives activity, even if they do not show up as headline AI winners. By contrast, capital-intensive story names with distant payoffs get treated more like long-duration options; when investors become selective, cash burn matters more than addressable market and the discount rate rises fast.

The contrarian miss is that broader breadth does not require equal performance. Dispersion can stay high, which argues for relative-value exposure rather than naked beta. The cleanest falsifier for the leader thesis is a second straight quarter of softer enterprise demand or margin compression; the cleanest falsifier for the speculative name short is evidence that runway has extended without proportional spending growth.

Key Decisions for Investors

• Buy NVDA on 3-5% pullbacks or via a 3-6 month call spread; target 8-12% upside as the market re-rates platform optionality, with a hard stop if hyperscaler capex commentary softens across two consecutive prints.
• Short SPCX into strength, or use 3-6 month puts if borrow/liquidity is workable; target 15-25% downside if investors refocus on burn and funding needs, and cover on any material financing, contract, or runway-extension surprise.
• Watch NDAQ as a quieter breadth beneficiary: initiate only if AI leadership expands into broader market participation and equity/options volume trends up for 2-3 weeks; use it as a lower-beta long rather than a chase trade.