According to Apollo Global Management's chief economist Torsten Slok, in 2026, the traditional 60/40 portfolio allocation—60% stocks for returns and 40% bonds for stability—has been replaced by an AI vs. non-AI split. The S&P 500's top ten holdings now account for approximately 40% of the index, with nine of the ten companies heavily AI-related, including Nvidia (7.5%), Apple (6.8%), Alphabet (6.4%), Microsoft (4.2%), Amazon (3.9%), Broadcom (2.8%), Meta (2.5%), Tesla (2.3%), and Micron (1.6%).
Capital allocation is heavily skewed toward AI across all markets: 87% of 2026 venture capital investment flows to AI enterprises, while 49% of new investment-grade corporate bond issuances and 38% of high-yield bond issuances are AI-infrastructure-related. AI data center investment accounts for approximately half of the U.S. GDP growth projected at 2% in 2026. The key risk lies in whether AI benefits will spread to the broader economy beyond the semiconductor and data center equipment sectors, with a significant slowdown in AI capex potentially cascading through the entire economy.