Bloom Energy (BE) is presented as a high-risk, high-reward investment targeting the electricity bottleneck created by the AI boom. The core thesis is that BE's fuel cells provide a more reliable and cost-effective power solution for data centers, a key advantage the market misunderstands. Validation from major customers like Oracle (ORCL) suggests growing adoption for this "picks and shovels" play on AI infrastructure. While highly volatile, some bull case models project a potential future stock price of $500-$800+ by 2030. This is considered a long-term investment for those with a high risk tolerance and a timeframe of 12 months or more.

Bloom Energy (BE)

• The Core Thesis: The primary bottleneck for the AI boom is not a shortage of GPUs, but a shortage of electricity and power infrastructure. Data centers require massive amounts of reliable, 24/7 power.
• The Technology: Bloom Energy provides fuel cells that generate electricity through a chemical reaction with natural gas, rather than combustion.
  • Reliability: They boast 99.999% uptime, equating to only minutes or a few hours of downtime per year, compared to gas turbines which can be down for 1-2 hours per week. This reduces the need for costly redundant systems.
  • Efficiency & Cost: While the upfront CapEx is 20-30% more expensive than traditional gas turbines, the "all-in" cost is 10-20% cheaper when factoring in several key savings:
    • Cooling Savings: The fuel cells produce 400°C exhaust heat which is used to power absorption chillers. This reduces the extra power needed for cooling from ~30% down to 5-10%.
    • Direct Power: The cells output 800-volt DC power, which is exactly what NVIDIA's next-generation racks require. This eliminates the need for costly transformers, UPS battery farms, and AC-to-DC conversion steps.
    • Oracle estimates they save $1.5 million per megawatt in CapEx due to these efficiencies.
  • Innovation Cycle: Unlike gas turbines, which have seen little change in 30 years, Bloom's fuel cell technology is constantly improving and becoming more efficient, similar to a technology or chip company.
• The Bull Case:
  • The company has seemingly stumbled into the perfect use case for its 20-year-old technology with the AI data center boom.
  • A bull case model suggests that if Bloom captures a fraction of the new AI-driven energy market, it could reach $20-$40 in earnings per share (EPS) by 2030.
  • Applying a standard 25-30x earnings multiple for mission-critical infrastructure could imply a potential future stock price of $500 to $800+. This represents a potential 5x to 6x return from current levels.
  • The hosts believe the company's growth is only capped by its ability to manufacture its "Bloom boxes."
• The Bear Case & Risks:
  • Poor Track Record: The company has a 20-year history of disappointing investors and missing expectations, leading to significant market skepticism.
  • AI Bubble: A general downturn in the AI sector would negatively impact demand.
  • Scandium Shortage: A short-seller report claims the company cannot scale due to a shortage of a required rare earth metal, Scandium. The hosts counter that this report is based on old, debunked data and that Bloom has been stockpiling the metal and is securing new supply lines.
  • Competition: New technologies like space-based data centers are mentioned, but the hosts believe they are decades away from being a realistic threat. Data center operators are conservative and prefer "decade-old" proven technology.
  • Stock Has Already Run: The stock is up significantly since the summer, but the hosts argue this was a simple repricing for solvency and does not factor in the massive future AI energy cycle.

Takeaways

• Bloom Energy is presented as a high-risk, high-reward, misunderstood "picks and shovels" play on the AI revolution, focusing on the critical bottleneck of energy.
• The investment thesis hinges on the idea that the market misunderstands the "all-in" cost-effectiveness and technical advantages of Bloom's fuel cells for AI data centers.
• The hosts have taken a large, long-term equity position (not options, due to volatility) and see it as a potential 12-month or longer trade.
• This is a volatile stock that is "not for the faint of heart." Investors should expect large price swings.
• The podcast strongly advises investors to do their own deep research into the bull and bear cases to determine if the opportunity is real or based on fear and misunderstanding.

NVIDIA (NVDA)

• NVIDIA's CEO, Jensen Huang, has publicly stated that energy is the primary bottleneck for future AI development, not GPUs.
• The company's next-generation chip architecture is designed to run on 800-volt DC power.
• This aligns perfectly with the power that Bloom Energy's fuel cells produce directly, creating a circumstantial and powerful synergy.
• The hosts speculate that there is a "high likelihood" of a future partnership or even an investment from NVIDIA into Bloom Energy because solving the power problem helps NVIDIA sell more of its high-end racks.

Takeaways

• The discussion reinforces the long-term demand for NVIDIA's products but highlights a critical dependency: the availability of power.
• An investment in a company like Bloom Energy could be seen as an indirect way to play the growth of NVIDIA, by investing in a key enabler of its ecosystem.

Oracle (ORCL)

• Oracle is the first major hyperscaler to aggressively adopt Bloom Energy's technology for its data centers, providing crucial validation for the technology.
• The company made a significant strategic investment in Bloom Energy, purchasing $480 million worth of stock at a $108 conversion price.
• Other hyperscalers are reportedly watching the Oracle/OpenAI deployment closely before committing, making Oracle's success a key catalyst for Bloom.
• Bloom Energy's stock has recently been trading down in correlation with Oracle's stock, as the market links their fates.

Takeaways

• Oracle's adoption is the most significant proof point for Bloom Energy's business case.
• Investors in Bloom Energy should monitor the progress of Oracle's data center build-out and any public statements they make about the technology's performance.
• The current negative sentiment around Oracle's stock may be creating a drag on Bloom Energy's price, potentially offering a buying opportunity if you believe in the long-term Bloom thesis.

Terran Orbital Corp (TAC)

• The investment thesis is that AI data centers are increasingly being built in Canada, and TAC is positioned to be a primary beneficiary of this trend.
• The hosts describe it as a "junkie company" with a poor track record, acknowledging the high level of risk.
• The company has delayed an announcement for a key partner in Canada, but the hosts expect news within the next few months.
• The stock was trading in the $14 area, which the hosts view as a potential buying opportunity.

Takeaways

• TAC is a speculative, high-risk AI infrastructure play based on a simple geographic theme.
• The investment depends on the company executing on its Canadian data center plans, which is uncertain given its "bad track record."
• The thesis has not changed for the hosts, who are holding their positions and would consider adding more at current levels.

Sphere Entertainment Co. (SPHR)

• The stock was trading at $91 a share, having run up significantly. One of the hosts sold at $78 and expressed regret.
• The business appears strong, with tickets for events like the Backstreet Boys costing as much as $1,300 per ticket.
• A major rumor is that Sphere might secure a Star Wars experience.
• The hosts speculate that if the Star Wars deal happens, the stock could go to $150 per share.

Takeaways

• SPHR has been a very successful trade, but there may be more upside.
• The primary future catalyst to watch for is the rumor of a Star Wars attraction, which could cause another significant move up in the stock price.