Paul Tudor Jones says AI bull market has 'another year or two to run'

Paul Tudor Jones said the AI-fueled bull market still has further to run and estimated the current phase is only 50-60% complete, implying another 1-2 years of upside. He added that he recently increased AI-related basket purchases, comparing today’s environment to the early Microsoft era and the mid-1990s internet acceleration. The comments are supportive for AI and megacap tech sentiment, but they are high-level macro views rather than a company-specific catalyst.

Analysis

The market implication is less about one man’s opinion and more about reinforcement: a prominent macro allocator publicly validating that AI remains a multi-quarter capital-rotation trade keeps under-allocated institutions chasing the same narrow set of winners. That matters because the marginal buyer is no longer just long-only growth funds; it is also systematic and macro capital that tends to scale into strength, which can prolong upside well past what fundamentals alone would justify. In that setup, the biggest beneficiary is not just AI software, but the full capex stack tied to compute, power, and network buildout, with second-order winners in utilities, grid equipment, and datacenter real estate.

The more interesting read-through is that consensus is likely underpricing the duration of earnings leverage rather than the direction of the theme. If the cycle is still in the middle innings, the next leg is usually driven by capacity constraints and pricing power in the supply chain, not just multiple expansion on the headline names. That creates a richer opportunity set in “picks-and-shovels” beneficiaries than in the most crowded mega-cap AI names, where good news is increasingly already embedded and disappointment risk rises if spending slows even modestly.

The main risk is a sentiment air pocket, not an AI thesis break. A 5-10% drawdown in the semiconductor complex could happen quickly if capex commentary slips, export restrictions tighten, or rates back up enough to pressure long-duration growth multiples. But over a 6-18 month horizon, the more probable reversal condition is not valuation alone; it is evidence that incremental AI spend is no longer translating into customer adoption or margin expansion, which would first show up in guidance cuts from infrastructure suppliers before it hits the index leaders.

Contrarian takeaway: the trade is probably over-owned in the obvious names but still under-owned in the enabling infrastructure. That argues for expressing bullish AI through balance-sheet-light suppliers and power-constrained beneficiaries rather than chasing the most celebrated platform stocks at peak narrative intensity. The risk/reward is better where the market still has to revise earnings estimates upward, not where it has already capitalized the story.

Key Decisions for Investors

• Stay long MSFT on a 6-12 month horizon, but use it as a lower-beta core holding rather than the highest-conviction expression of the AI trade; upside is likely steadier than explosive, with a favorable risk/reward if enterprise AI monetization keeps compounding.
• Rotate incremental AI exposure out of the most crowded mega-cap basket and into AI infrastructure suppliers over the next 1-3 months; target names with operating leverage to capex and pricing power, where estimate revisions are still ahead of the stock.
• Pair trade: long AI power/grid beneficiaries vs. short a basket of high-multiple software names that depend on future AI adoption acceleration; if the narrative pauses, the short leg should compress faster than the infrastructure longs re-rate.
• Buy 6-12 month call spreads on semis or AI infrastructure on any 5-7% sector pullback; this preserves upside if the cycle extends while defining risk if sentiment rolls over.
• Set a tactical trim level for crowded AI winners after 10-15% additional upside from current levels unless there is clear evidence of accelerating orders; the asymmetry worsens once positioning gets fully consensus.