What’s Going On With AI Stocks? | The Brainstorm EP 110

Key Concepts

• Macroeconomic Factors: Government shutdowns, Federal Reserve (Fed) interest rate decisions, and their impact on market liquidity.
• AI Bubble Narrative: Debate on whether the AI market is a bubble, with arguments focusing on demand, monetization, and overvaluation.
• Nvidia: Discussion of Nvidia's role in the AI boom and the significance of major investors selling their stakes.
• Monetization Lag: The concept that revenue generation often follows technological adoption and productivity gains.
• AI Company Monetization Strategies: Differences between companies with existing revenue streams (e.g., Google/Alphabet) and those relying on new products/subscriptions (e.g., OpenAI).
• Price Competition in AI: The potential for large tech companies to use low pricing to stifle competition from newer AI firms.
• Switching Costs: The difficulty or expense for users to transition from one AI platform to another, especially with embedded context and personalization.
• Crypto Market Dynamics: Factors influencing Bitcoin's sell-off, including liquidity, leverage, and the impact of institutional investors and ETFs.
• Reusable Rockets: The progress and competition in the reusable rocket technology sector, particularly between SpaceX and Blue Origin.
• Prediction Markets: The use of prediction markets for forecasting events and their role in price discovery and risk management.

Market Downturn and AI Bubble Debate

The current market downturn is attributed to several macroeconomic factors, including a government shutdown impacting liquidity and the Federal Reserve's upcoming interest rate decisions, which are being made without a complete set of macroeconomic indicators. Fed futures show a decreased probability of a rate cut.

A prevailing narrative suggests that the AI market, particularly Nvidia, might be a bubble that is now popping. This perspective posits that increased spending on data centers will become wasteful as demand falters. However, the speakers largely disagree with this assessment.

Arguments Against an AI Bubble

• Sustained Demand: The core argument against an AI bubble is that demand for AI tools will not deflate as long as users continue to pay for them. The productivity gains delivered by AI are expected to triple this year and potentially triple again next year. This increasing productivity justifies a higher willingness to pay, even if it's not a 9x increase over two years.
• Massive User Adoption: Projections indicate billions of users for AI chatbots by the late 2020s, possibly as early as 2028, suggesting a strong and growing user base.
• Pricing Support: This user adoption translates to significant pricing support, estimated at 40-50x.
• Supply-Side Bottlenecks: Any perceived slowdown in AI chip deployment might be due to supply-side issues (e.g., lack of power outlets for data centers) rather than a lack of demand.

Key Events Fueling the Bubble Narrative

• Peter Thiel's Nvidia Stake Sale: Peter Thiel reportedly sold his entire Nvidia stake.
• SoftBank's Nvidia Stake Sale: SoftBank also disclosed selling its Nvidia stake. These actions are being used to question Nvidia's valuation and the late stage of the AI rally.

Monetization Challenges and Competitive Landscape in AI

A significant point of discussion is the lag between AI adoption and monetization, and how this plays out in a market heavily concentrated on AI.

Monetization Strategies and Disagreements

• Two Types of AI Companies:
  • Existing Revenue Streams: Companies like Google (Alphabet) leverage existing business lines (e.g., advertising) to fund AI development.
  • New Products/Subscriptions: Companies like OpenAI rely on new products and subscriptions for revenue.
• Disagreement on Pricing Power:
  • Brett's View: Believes there's a significant gap between what users are willing to pay and current prices, suggesting companies like OpenAI have room to increase pricing.
  • Nick's View: Argues that companies like Alphabet are in a strong position to keep AI pricing low to squeeze competitors like OpenAI. Alphabet's advertising revenue provides substantial cash flow, incentivizing them to maintain low prices to maintain market share and lock users into their ecosystem.

Google's Strategic Advantage

• Bundling and Ecosystem Integration: Google can bundle AI services (like Gemini) across its products (Chrome, GCP), creating a sticky ecosystem that is difficult for standalone AI companies to replicate.
• Artificial Price Suppression: The argument is that Google might artificially keep AI prices low, not necessarily to maximize immediate AI revenue, but to gain market share and reinforce its core businesses. This could reduce their return on capital for AI specifically but strengthen their overall position.
• User Engagement Metrics: While ChatGPT users spend around 15 minutes daily, Gemini users spend significantly less (around 2 minutes). This suggests that while AI tools are valuable, the current engagement patterns might not fully translate to sustained high willingness to pay for standalone applications, especially if they are perceived as novelties rather than essential tools.

Counterarguments to Google's Dominance

• Destructive Margins: Aggressively competing on price could significantly impact Google's margins, potentially affecting their capital expenditure plans, similar to Meta's experience with its metaverse investments.
• Unmet Demand: If OpenAI and other AI companies cannot meet demand due to GPU shortages, Google's incentive to price low diminishes. Offering free services like image generation (e.g., Google's Nano Banana) might be a way to service users while OpenAI focuses on paying customers.
• Enterprise Business Growth: OpenAI's enterprise business, which is 100% monetized, is growing faster than its consumer business, indicating strong demand for paid AI solutions.

Switching Costs and User Stickiness

The discussion delves into the concept of switching costs for AI users.

• Brett's Perspective (High Switching Costs): Brett argues that switching AI platforms is becoming increasingly difficult due to the embedded context, personalized settings, and learned workflows. He uses personal examples like managing family health information or curating book recommendations for his children, where the AI's accumulated knowledge creates significant switching friction. He likens it to switching firms or major life changes.
• Nick's Perspective (Lower Switching Costs for Consumers): Nick believes that for the average consumer, switching costs are currently low. He uses Gemini, Perplexity, and ChatGPT interchangeably, suggesting a lack of deep integration or personalization that would make switching painful. He also points out that Google can leverage its existing services (Gmail, etc.) to reduce switching costs for users.
• The Future of Switching Costs: While acknowledging that enterprise adoption might lead to higher switching costs due to data integration, the debate continues on whether consumer-level personalization will create similar barriers. The rapid improvement of AI models (e.g., 20% productivity gain now, projected 60-70% in a year) could increase the perceived value and thus switching costs over time.

Crypto Market Analysis

The crypto market is experiencing a sell-off, with explanations focusing on liquidity, leverage, and a potential shift in market cycles.

• Liquidity Drain: Macroeconomic factors like government shutdowns are draining liquidity from the system.
• Leverage and Liquidations: There's a suspicion of a hedge fund blow-up due to excessive leverage, possibly in Asia. This feels specific to native crypto markets rather than traditional finance.
• Dormant Coin Movement: On-chain data shows dormant Bitcoin coins being moved to market, suggesting core crypto holders might be taking profits.
• Four-Year Cycle and Leverage: The expectation of a four-year cycle and parabolic moves has led some to take on leverage, which is now being unwound as the market moves in the direction of maximum pain for leveraged positions.
• ETF Impact and Gatekeepers: The introduction of Bitcoin ETFs has brought in a larger pool of money, but the gatekeepers (financial advisors) are not yet standardly recommending these products. This means the "giant lump of money" is not yet fully flowing into crypto, potentially extending the cycle beyond what crypto natives expected.
• "Climbing the Wall of Worry": The sentiment is that both bulls and bears wanted a blow-off top, but the market might instead "climb the wall of worry" as expectations are reset.

Reusable Rockets and Space Exploration

The recent successful landing of a rocket by Blue Origin, nearly 10 years after SpaceX's first landing, is discussed.

• Blue Origin's Achievement: Blue Origin successfully landed a rocket, a significant milestone that took a decade to achieve.
• SpaceX's Lead: Despite Blue Origin's success, SpaceX is perceived to be significantly ahead, particularly with its Starship program, which aims for full reusability and potentially an order of magnitude reduction in launch costs.
• Pace of Innovation: The discussion highlights the difference in the pace of innovation between the two companies, with SpaceX continuously pushing boundaries while Blue Origin appears to be catching up.
• Emerging Players: Rocket Lab and Stoke Space are mentioned as other companies working on reusable rocket technology. Rocket Lab is expected to achieve reusability in early 2026.
• Capitalism and Progress: The achievement is framed as a testament to capitalism's ability to drive progress and innovation in space exploration.

Prediction Markets and Gemini 3.0 Release

The conversation touches upon prediction markets and the anticipated release of Gemini 3.0.

• Gemini 3.0 Release Odds: A prediction market shows a 66% chance of Gemini 3.0 being released tomorrow (Tuesday, November 18th), with decreasing odds for subsequent days.
• Alphabet CEO's Engagement: Alphabet CEO Sundar Pichai retweeted the prediction market with an emoji, suggesting awareness and potential influence on the release timing.
• Prediction Market Anecdote: A solar developer used a prediction market to assess the true value of a tariff refund opportunity, enabling them to negotiate better terms with a bank that was trying to offer a significantly lower price. This highlights how prediction markets can provide price discovery and cut out intermediaries.
• Market Efficiency: The discussion questions the efficiency of prediction markets, noting that while they are generally accurate, the volume of trading might be low. The need for higher volume to facilitate risk offloading and broader market participation is acknowledged.

Conclusion

The speakers largely agree that the current market downturn is not indicative of a full-blown AI bubble but rather a healthy pullback. While there are valid concerns about competition and monetization strategies among AI companies, the underlying demand and productivity gains suggest continued growth. The crypto market is seen as experiencing a natural correction driven by liquidity and leverage unwinding, with a potentially longer cycle than anticipated. In the space sector, while Blue Origin has achieved a significant milestone, SpaceX maintains a substantial lead in reusable rocket technology. Prediction markets are emerging as valuable tools for price discovery and risk assessment.