Citrini: The Next Phase of The Stock Market's AI-Obsession

Episode Summary

Executive Summary: Citrini argues that AI remains a real fundamental trend, but the easy money in broad picks-and-shovels trades has largely been made. He sees a more selective phase ahead, with winners shifting toward memory, optics, custom silicon, and AI adopters, while macro risks, elections, and credit stress could influence positioning. He also discusses his risk management, basket-based trading process, and selective views on China, the S&P 500, and fiscal beneficiaries.

Main Topics: AI as a real trend, but selective phase ahead (Priority: 5/5): Citrini believes AI is not مجرد hype: revenue and earnings gains justify much of the move. However, he thinks the market is moving from broad exposure to a more selective environment where picking the right sub-industries and end beneficiaries matters more. Bubble dynamics and the Soros/Gartner framework (Priority: 5/5): He frames bull markets through a bubble cycle: early long, late short, and later re-entry at the trough. He thinks AI has not yet reached dot-com-style excess, but estimates and narratives could become disconnected later. Magnificent Seven, positioning, and the S&P 500 (Priority: 4/5): He is constructive on the AI-linked parts of the mega-cap complex but less excited about the broader index, which he expects to be choppy and potentially flat. He emphasizes that positioning squeezes helped drive the rally, but that effect is now largely gone. Macro hedging, rates, and credit risk (Priority: 4/5): Citrini is willing to hedge cyclical exposure with rates trades when asymmetry appears favorable. He sees more concern about recession/credit stress than inflation reacceleration, and worries about loan-loss provisions and banking stress. China trade: successful early, then more difficult (Priority: 4/5): He explains his long India/short China trade, subsequent China reversal, and current China basket approach. He views China as a macro-driven market with heavy policy risk and weak stock-specific alpha, despite pockets of opportunity. Fiscal beneficiaries and election-driven themes (Priority: 3/5): He is building baskets around higher structural fiscal spending, grid/electrification, infrastructure, and aging demographics. He believes the election can shift sector winners even if it doesn't radically change index-level returns. Research process, expert calls, and risk management (Priority: 5/5): He describes a generalist process: identify a theme, speak with domain experts, map beneficiaries and losers, then size and hedge the trade carefully. He stresses stop discipline, concentration limits, and scaling out only when the portfolio demands it.

Key Arguments: AI is fundamentally real because revenue and earnings are rising alongside stock prices, unlike much of the 2020-2021 speculative boom. The market’s first easy AI trade was the broad picks-and-shovels basket; future alpha will require more granular work on memory, optics, custom silicon, and actual adopters. Positioning mattered early because many investors were underweight AI winners; that squeeze is now mostly over. The S&P 500 may be broadly boring and range-bound even if AI-related leaders remain strong. Macro risks now skew more toward recession/credit problems than a renewed inflation surge, making duration/bond hedges more attractive at times. China remains investable only as a macro, policy-sensitive trade; company-specific alpha is hard because policy can change quickly. Good trading is about asymmetric risk, strict stops, and not overriding your plan just because a position has moved partway in your favor. A theme only works when there is both a real catalyst and a favorable macro backdrop; demographics or water alone are not enough. Fiscal spending themes can persist across administrations, but sector winners may change depending on policy priorities. If AI turns into a true mania, the best-underwritten quality names may underperform the most speculative, narrative-driven stocks despite strong fundamentals.

Data Points: Super Micro share price: about $575 - Discussed as the current level after a major rally since Citrini first bought around $56 in August 2022. Super Micro gain since last interview: up 138% - Performance since the June 2023 prior interview. Fabrinet gain since last interview: up 85% - Referenced as one of the names he had been bullish on. Elf Beauty gain since last interview: up 51% - Referenced as part of his successful stock calls. NVIDIA gain since last interview: up 56% - Used to illustrate AI leadership since the previous interview. Super Micro first purchase price: $56 - He said his first shares were bought in August 2022 at this level. Super Micro forward P/E: about 26x - He argued it was still cheaper than Chipotle on a forward earnings basis. Super Micro projected revenue: $25 billion annual revenue in the next four years - His estimate of where the business could reach. UBS AI industry revenue forecast: $420 billion - He cited UBS predicting AI industry revenue, saying it was a 40% increase from prior estimates. UBS forecast growth rate: 72% CAGR from 2022 over five years - He cited this as still achievable, and even conservative by bubble standards. NVIDIA revenue growth: up 60% year over year - Used to argue fundamentals were supporting the stock move. NVIDIA net income growth: up 800% in Q3 - Cited as evidence of real profitability, though he noted the base effect. Bonds trade timing: January 19 - He said he called a bond bottom for the year on January 19 and bought call options. Bond move after his call: rates moved 40 basis points - He said the trade worked after his call. China basket drawdown: down about 7% - Current loss on his China basket at the time of the interview. China stop-loss threshold: 10% drawdown - He said he would exit if the basket goes below this level. Portfolio concentration limit example: 300 basis points - He mentioned Super Micro as a large position he scaled due to concentration limits. S&P 500 outlook: plus or minus 5% - He predicted the index could end the year roughly flat within this range.

Pivotal Quotes: "My loyalties are to the number on my screen, right? My P&L." — Citrini: He explains that he is not a true believer in narratives and that his focus is on price and profitability. "Take the risk that you're paid for and hedge the rest." — Citrini: He uses this trading principle to explain why he hedges cyclical exposure while staying long AI exposure. "You want to be early, but you don't want to be alone." — Citrini: He describes how he times thematic trades: early enough to catch the trend, but with confirming catalysts and macro support.

Implications: Listeners should expect a more selective AI market, with gains shifting from broad exposure to narrower sub-sectors and winners. Macro hedging, fiscal policy, and credit conditions may matter more for returns than simple direction calls.