The Magnificent 7 Is Dead. These 10 Stocks Are the New Market Favorites

The article argues that the market leadership framework is expanding beyond the Magnificent 7, with a new 'FAB 10' group adding OpenAI, SpaceX, and Anthropic to reflect the AI era. It highlights Nvidia at $4.95 trillion, Alphabet at $4.40 trillion, Apple at $4.35 trillion, and SpaceX at $2.53 trillion, while Broadcom, TSMC, and memory chip makers are becoming major AI beneficiaries. The tone is constructive for AI and semiconductor exposure, but the piece is primarily commentary on shifting investor positioning rather than a single price-moving event.

Analysis

The market is transitioning from a narrow index-driver regime to a broader capital-intensity regime. That matters because the second-order winners are no longer just the model builders; they are the bottleneck owners in the AI stack — advanced foundry capacity, HBM memory, optical interconnects, and AI networking silicon. In practice, that shifts incremental economic value away from the brand-layer hyperscalers toward the infrastructure layer where supply remains constrained and pricing power is highest.

The near-term risk is that investors may overpay for the “new leaders” narrative while underestimating how cyclical some of these beneficiaries remain. Memory and foundry exposure can be excellent trades over 6-18 months, but they are still exposed to capex digestion and eventual supply response; once lead times normalize, margin expansion can compress quickly. The more durable winners are the picks-and-shovels names with embedded design wins and multi-year sockets, not the names simply riding sentiment or a one-time re-rating.

A key contrarian read is that private-market halo effects can create a false sense of scarcity. The rush into frontier AI private names may be less about fundamentals than about investors seeking exposure they can’t get in public markets, which can pull valuation multiples above what public comps justify. That suggests the public-market trade should be selective: prefer bottleneck suppliers and defense-linked infrastructure over generic AI-adjacent software, where the monetization path is farther out and easier to disappoint.

The positioning implication is that the old Magnificent 7 basket is likely to underperform on a relative basis only if capital rotates into under-owned enablers rather than simply broadening within mega-cap tech. The clearest path for alpha is to own the constraint, not the narrative, and to express that with pairs that isolate supply-chain leverage from duration risk in the broader AI theme.

Key Decisions for Investors

• Long AVGO vs. short a basket of high-multiple AI software proxies for 3-6 months; AVGO has direct exposure to AI networking bottlenecks and better near-term pricing power, while software names remain more execution-sensitive.
• Long TSM and MU on a 6-12 month horizon, but size as a cyclical trade rather than a secular compounder; reward is continued AI-driven utilization and memory tightness, risk is capex catch-up and multiple compression once supply normalizes.
• Maintain a tactical long NVDA, but hedge with a partial short in MSFT or AMZN if using a basket approach; this isolates the infrastructure spend wave while reducing exposure to any slowdown in hyperscaler capex pacing.
• Avoid chasing private-market AI enthusiasm via public proxies for the next 1-2 quarters; the better entry is after the first post-hype drawdown, when capital rotation creates a cheaper re-entry into the enablers.
• If forced to express the theme in one pair, go long TSM/AVGO and short TSLA; the former have clearer AI bottleneck economics, while TSLA’s sentiment premium is less tied to the core AI spend cycle described here.