Invest Like the Best with Patrick O'Shaughnessy

• For private-market investors, prioritize strong founders in overlooked, non-obvious markets, but scrutinize valuation, margins, customer concentration, and revenue quality; rapid growth alone is not enough.
• Treat AI compute infrastructure as a high-risk opportunity: demand is strong, but rising borrowing costs and heavy financing needs could undermine companies’ economics.
• Watch AI data providers for hard-to-replace data and durable customer demand, while checking whether low margins (around 15–20% for some labeling businesses) limit returns.
• Be cautious with vertical SaaS priced as high as 100× revenue; seek evidence that products actually replace labor before accepting the premium.
• Consider private-company examples like Conduit, New Limit, and Chai Discovery speculative watchlist names, not actionable buys; the discussion provides no entry prices or financial data.

• Consider Snowflake (SNOW) as an established data-infrastructure business that may remain resilient to AI disruption, while closely monitoring competition from Databricks; no price target was provided.
• Track AI networking as a potential infrastructure opportunity as larger AI clusters increase demand, but wait for evidence of durable growth before treating Arista Networks (ANET) as a buy.
• Approach AI infrastructure and neoclouds selectively: prioritize providers with secured chips, power, sites, and financing, since hardware and funding constraints could limit growth.
• Treat frontier AI model leadership as fragile—model advantages may last only three to six months—and watch for slower demand or delayed releases that could weaken pricing power.

• Meta (META) has a constructive AI thesis: monitor whether AI improves user engagement and ad performance; no price target was provided.
• Treat Carvana (CVNA) as a high-risk, concentrated investment—the guest’s roughly 80% portfolio stake reflects a personal conviction, not a broadly suitable allocation.
• Consider TSMC as a potential beneficiary of hard-to-replicate manufacturing expertise, while watching for competitors building alternative capacity.
• Be cautious about assuming AI spending guarantees supplier returns: the discussion highlighted overbuilding and commoditization risks for NVIDIA (NVDA) and other infrastructure providers.
• Grindr (GRND) is only a speculative AI product-expansion possibility, not an endorsed investment; no specific products or targets were identified.

• Consider cybersecurity as a long-term growth theme: AI may accelerate attacks and increase demand for faster detection and remediation, though no specific stock or entry price is recommended.
• Compare security providers by customer segment and economics: CrowdStrike (CRWD) is discussed in enterprise use cases, while Vanta targets a broader set of customers but remains private.
• Evaluate AI software companies for sustained customer adoption, workflow value, and durable data or integration advantages—not AI branding alone; the insights provide no specific public-market buy, price target, or timeframe.

• NU offers the clearest long-term opportunity, with 140 million customers and a low-cost digital banking model; consider it a growth investment, while monitoring customer growth, customer-acquisition costs, and profitability as it expands internationally.
• Treat Bitcoin (BTC) as a possible component of a diversified portfolio, not a short-term trade: no price target, allocation, or timeline was provided.
• Dollar stablecoins and digital banking are promising themes, but the insights identify no specific investment beyond NU or actionable price targets.

• Treat neurotechnology and AI-enabled brain-computer interfaces as long-term watchlist themes, not proven near-term investments; the discussion named no actionable public-company trade or catalyst.
• Monitor sleep and health-monitoring technology for products that demonstrate reliable benefits, but Eight Sleep, WHOOP, and Oura were examples—not recommendations.
• No specific ticker, price target, or timeframe was provided, so avoid buying solely on these speculative forecasts.

• Instinct is private, so it is not a directly investable public stock; treat its reported growth and transaction figures as unverified, and watch for evidence of durable revenue and manageable compute and security costs.
• The AI-agent theme remains speculative: monitor whether agents reliably generate transactions and whether public companies can benefit without losing customer relationships or economics.
• For Alphabet (GOOGL/GOOG), Meta (META), and Snap (SNAP), track changes in user engagement and ad exposure, but the discussion provides no basis for a specific trade.
• Monitor Amazon (AMZN), DoorDash (DASH), Uber (UBER), and Lyft (LYFT) for agent-driven ordering and booking adoption; easier access could boost transactions even as their apps become less central.

Investors should consider positions in major investment banks like JPMorgan Chase (JPM), Goldman Sachs (GS), and Bank of America (BAC), as they are primed to expand profit margins by automating deal execution and capturing new volume in smaller business transactions. In enterprise technology, focus capital on specialized vertical AI software companies that solve complex regulatory and workflow integrations rather than general foundation model developers. Target software vendors transitioning to outcome-based pricing models, which allow them to directly monetize completed transactions rather than basic software seats. Within alternative assets, prioritize private equity and private credit firms adopting automated due diligence tools that compress transaction turnaround times from months to hours. Additionally, track emerging financial software platforms building secondary market liquidity and structured data exchanges for private assets as high-upside infrastructure investments.

Investors should maintain long-term exposure to top-tier Artificial Intelligence (AI) and software leaders that are achieving massive scale and market dominance far faster than previous tech generations. Prioritize AI platforms driving broad productivity gains, focusing on long-term economic expansion rather than near-term labor disruption. Consider an investment in Xiaomi Corporation (1810.HK / XIACF) as the company successfully expands beyond consumer electronics to capture market share in the high-performance electric vehicle (EV) sector. Broaden your equity strategy to favor agile management teams with a proven track record of redeploying supply chain mastery and software expertise into large adjacent industries. Finally, for allocations in early-stage venture capital and private tech markets, direct capital toward exceptional founders who harness modern software tools for extreme operational leverage rather than traditional funds relying on disappearing information advantages.

Investors should maintain core exposure to U.S. Big Tech platforms like Alphabet Inc. (GOOGL), which are structurally insulated from physical trade barriers and positioned to drive productivity through artificial intelligence in enterprise and life sciences.

Within the defense sector, reallocate capital away from legacy heavy hardware toward software-driven defense technology, autonomous drones, and counter-drone systems built for modern, rapid-iteration warfare.

For international diversification, allocate to India as a leading long-term emerging markets play driven by demographic expansion and an accelerating domestic tech sector.

Real estate investors can capitalize on the ongoing shift to hybrid work by investing in suburban and exurban residential development and remote work infrastructure that expand traditional commuting zones.

Lastly, seek long-term exposure to frontier deep-tech leaders like SpaceX, OpenAI, and Anthropic, as computational and aerospace capabilities increasingly align with national strategic infrastructure.

Investors should build long-term positions in nuclear energy (specifically Small Modular Reactors) and natural gas infrastructure to solve data center power constraints ahead of the critical 2030 to 2032 bottleneck.

Reduce supply-chain risk by investing in domestic foundries and alternative specialized silicon accelerators to diversify away from single dependencies like Taiwan Semiconductor Manufacturing Company (TSMC).

Target computational biology and pharma AI platforms such as Chai Discovery, which are actively generating multi-million-dollar enterprise revenues by accelerating pharmaceutical drug discovery.

Prepare for near-term commercial breakthroughs in physical AI and semi-humanoid robotics by tracking innovators like Sunday Robotics ahead of consumer beta launches scheduled for late this year into early next year.

Allocate toward high-conviction vertical enterprise AI software, led by platforms like Harvey, that demonstrate strong enterprise pricing power by automating complex workflows in legal, accounting, and compliance sectors.

Investors should maintain core exposure to NVIDIA (NVDA) for premium real-time AI workloads and Taiwan Semiconductor Manufacturing Company (TSM) for indispensable advanced chip manufacturing leverage. Capitalize on critical hardware bottlenecks by investing in memory suppliers like Micron Technology (MU), SK Hynix, and Samsung, which stand to profit from persistent shortages in High Bandwidth Memory (HBM) and DRAM. Consider Advanced Micro Devices (AMD) as a compelling value play as major tech firms adopt cost-effective chip alternatives to diversify away from premium GPU pricing. Hold Intel Corporation (INTC) as a long-term strategic hedge against potential Asian semiconductor supply chain disruptions. Finally, expand into the emerging AI Inference Infrastructure and flexible power generation sectors as data center demand shifts toward low-cost, decentralized computing for background AI tasks.

Invest in Alphabet (GOOGL) and Meta Platforms (META) for immediate, high-margin returns from AI-optimized advertising, complemented by Amazon (AMZN), which lowers cloud costs through custom silicon like Trainium.

Accumulate Intel (INTC) to capture upside from its foundry turnaround as Big Tech actively diversifies away from severe Taiwan Semiconductor Manufacturing Company (TSM) supply shortages projected to last through 2028–2029.

Hold Apple (AAPL) as a low-risk AI distribution play that captures high-margin software integration on consumer hardware without incurring heavy data center buildout costs.

Exercise caution with NVIDIA (NVDA), as long-term gross margins face compression from custom hyperscaler chips and deferred balance-sheet liabilities tied to compute buybacks through 2030.

Monitor Microsoft (MSFT) closely, as high computing costs and friction from usage-based pricing on enterprise tiers (such as the $100 per user per month E7 tier) may compress traditional software margins.

Capitalize on the artificial intelligence boom by investing in market-leading hardware provider NVIDIA ($NVDA**), which powers the essential compute infrastructure for AI models. Look beyond pure AI software and invest in the dominant cloud oligopoly of Amazon Web Services ($AMZN), Microsoft Azure ($MSFT), and Google Cloud Platform ($GOOGL), as enterprise adoption accelerates. Take advantage of severe multiple compression in the Software-as-a-Service ($SaaS)** sector by selectively buying established enterprise software companies successfully pivoting to AI. Play the massive energy bottlenecks caused by surging data center demand by investing in power generation, electrical grid modernization, and nuclear energy infrastructure. Avoid zero-sum traps in emerging technologies and instead build a balanced portfolio across these interconnected semiconductor, cloud, software, and energy themes over the next 3 to 5 years.

NVIDIA (NVDA) currently trades at its cheapest forward P/E multiple in a decade, presenting a prime long-term entry point for investors despite recent AI sector volatility. Underlying fundamentals for NVIDIA (NVDA) remain exceptionally strong as token growth and GPU rental pricing continue to accelerate rapidly. Investors should capitalize on short-term market panics to buy top-tier hardware and infrastructure providers like Microsoft (MSFT), Meta (META), and Amazon (AMZN), whose hyperscaler cash flows are surging. As older contracted compute bases expire, upcoming price re-adjustments will significantly drive revenue acceleration across the semiconductor sector.

To capitalize on the booming artificial intelligence sector, focus your investments on foundational infrastructure providers rather than speculative application startups. Microsoft (MSFT) and **Oracle (ORCL) **remain top choices for capturing surging enterprise cloud demand and hosting large-scale AI operations. Meanwhile, NVIDIA (NVDA) sits at a critical supply chain bottleneck, providing the essential advanced hardware and GPUs required for all modern AI workloads. Investors should also target specialized chip designers and energy providers that drive crucial efficiency gains as data center power demands skyrocket. Always monitor potential risks related to permitting challenges and public pushback regarding the massive physical footprint of these new facilities.

Investors should prepare for a structural Natural Gas supply deficit starting in 2028, which could drive prices from the current $3.50 range to over $8.00 per MCF. Expand Energy (EXE) is a high-conviction play currently trading at a low 4x EBITDA, offering significant upside due to its control of critical Haynesville shale inventory. For exposure to the Appalachian region, EQT Corporation (EQT) and Range Resources (RRC) are top-tier producers positioned to benefit as gas becomes the primary fuel for U.S. power generation. To hedge against rising electricity costs, look toward utility-scale solar providers like NextEra Energy Partners (NEP) and Clearway Energy (CWEN), which can reprice power contracts at much higher margins. In the nuclear sector, Cameco (CCJ) and BWX Technologies (BWXT) are the strongest long-term bets to benefit from the necessary expansion of large-scale reactor infrastructure.

Focus on high-growth infrastructure providers like WorkOS, Ramp, and Vanta, which are becoming essential for enterprise AI adoption and automated compliance. Monitor the Defense Tech sector closely, but prioritize firms with a long-term track record over those recently pivoting to the trend to avoid "diluted differentiation." When evaluating new ventures like Lila Sciences, apply the Persuasion Equation by ensuring the team has a clear plan to mitigate risk while driving high desire through "scientific superintelligence." Look for investment opportunities that exhibit true Scarcity, similar to the Benchmark model, as these assets move capital the fastest and maintain the best terms. Prioritize companies with a simple, defensible value proposition, as high Complexity in a business story often signals a lack of reliability and slows down institutional support.

Investors should target "Old" Private Companies (6–7 years old) that have recently inflected due to AI catalysts, as their slow starts often lead to significant undervaluation compared to newer startups. Avoid traditional SaaS companies with per-seat pricing models and instead prioritize AI-native software firms that utilize usage-based pricing to offset higher compute costs. For private market exposure, prioritize Emerging Managers over large growth funds, as smaller managers are more psychologically and financially aligned with generating high returns rather than collecting fees. Retail investors should exploit their lack of "career risk" by holding high-conviction, "weird" assets like Bitcoin or "Long Elon" plays (Tesla, SpaceX secondary markets) that institutional mandates often restrict. Monitor X (formerly Twitter) as a leading indicator for market sentiment, as algorithmic narratives now drive high-variance pricing in the Mag 7 and broader tech sectors.

The shift from AI training to inference (running models) is creating a massive opportunity for specialized hardware that outperforms general-purpose GPUs. Investors should look toward TSMC (TSM), which remains the essential manufacturing partner for next-generation AI chips like those from Etched and big-tech hyperscalers. While NVIDIA (NVDA) remains the industry benchmark, its high margins are under threat from application-specific chips (ASICs) designed to run "Transformer" architectures more efficiently. Consider exposure to the "Token Economy" by focusing on companies that maximize "tokens per watt," as energy availability is now the primary bottleneck for AI scaling. By 2027, the market is expected to pivot from "per-seat" software pricing to "per-inference" costs, benefiting high-growth AI agent platforms like Perplexity and Anthropic.