Brad Gerstner: No AI Bubble, Semis Eat the Nasdaq & AI's Take Off Problem
Summary
Brad Gerstner joins the show to discuss the state of the AI market, emphasizing that current valuations are driven by earnings rather than multiple expansion. He highlights the critical need for AI labs to hit specific revenue targets to justify massive infrastructure capex, while noting risks from potential interest rate hikes and regulatory challenges. Gerstner predicts a slower compute buildout than forecasted and advises investors to remain flexible, focusing on monthly revenue data and macroeconomic factors to navigate the next phase of the AI supercycle.
Topics discussed
Introduction of Brad Gerstner and Trump Accounts
Heart health initiatives and C-scan advocacy
Market overview: Performance despite macro concerns
Earnings-driven expansion and valuation metrics
Sector performance: Semiconductors vs. laggards
Hyperscaler capex and semiconductor cash flow
AI revenue milestones: Anthropic and OpenAI
Market consolidation and open source concerns
Parabolic revenue curves and historical context
Capex funding gap and off-take revenue needs
Compute expansion forecasts for 2026-2028
Total Addressable Market (TAM) for knowledge work
Token growth and enterprise AI adoption
Productivity dividends and margin expansion
Regulatory risks and pragmatic solutions
Historical parallels: Nuclear energy and regulation
Infrastructure challenges and compute build-out
Interest rate hikes and cost of capital
Portfolio strategy and fan of potential outcomes
2026 outlook: Flexibility and risk management
Listen ad-free on Castria