• Consider Reddit (RDDT) as a volatile, moderate-sized position: its fresh human discussions may gain value through AI-data licensing, but user growth has slowed and future licensing increases are uncertain.
• For long-term AI infrastructure exposure, Amazon (AMZN) is a higher-conviction choice, with potential upside from AWS and its AI chips; keep leverage low and account for possible pressure on advertising.
• Bloom Energy (BE) offers a higher-risk way to invest in rising data-center power demand, but its roughly 12% daily swings argue for position sizing carefully; no entry target was provided.

Reddit (RDDT)

• Chris described Reddit as a valuable source of fresh, human-generated data, especially discussions that show people arguing, explaining their views, and sharing experiences.
• He said Reddit’s user growth had slowed to 7% in the latest reporting cycle, while the company said 5% of all content ever written on Reddit was created in the most recent quarter. The hosts could not determine how much of that new content was AI-generated.
• Reddit makes data available to researchers for free but expects commercial users, including AI labs, to pay for commercial use. The hosts speculated that licensing deals could become more valuable when renewed, though they did not confirm that future price increases are assured.
• Meta’s newly announced Reddit competitor triggered a brief decline in Reddit’s share price. The hosts generally doubted the threat, arguing that Reddit is organized around shared interests and that Meta has struggled to establish similar discussion platforms.
• Risks raised in the discussion include AI-generated content on Reddit, slower user growth, uncertainty about whether licensing revenue will rise, and unpredictable market reactions.

Takeaways

• The transcript’s clearest bullish thesis is that Reddit’s human discussions could become more valuable as AI companies seek current, authentic training data. Chris said he bought more shares and characterized his position as moderate.
• Treat future licensing growth or an acquisition as speculation, not a certainty. The hosts also described Reddit as volatile and cautioned against overconfidence in short-term market moves.

Amazon (AMZN)

• Chris said he remains long-term bullish on Amazon, particularly because agentic AI could increase demand for AWS computing.
• He also pointed to Amazon’s Bedrock, Graviton, and Trainium businesses, as well as the company’s work on an agentic version of Alexa called Project Moonraker.
• A potential concern is that AI agents making purchase decisions could weaken Amazon’s advertising business. The hosts also noted that Amazon has time to adapt and could develop new ways for sellers to reach customers through agents.
• The discussion mentioned possible pressure from diesel prices and shipping costs. Chris also cautioned that near-term news and market reactions are unusually difficult to predict.

Takeaways

• The discussion presents Amazon’s cloud and computing businesses as potential beneficiaries of AI, while identifying advertising as a possible area of disruption from agentic shopping.
• Chris expressed confidence in Amazon’s long-term outlook but urged investors to keep leverage under control and avoid strong short-term assumptions.

Bloom Energy (BE)

• Bloom Energy’s stock was described as highly volatile, with the hosts noting moves of roughly 12% up or down on different days.
• One host mentioned that Bloom was expanding its factory and fuel capacity. The broader argument was that electricity suppliers and energy infrastructure may benefit as AI requires more power.
• The hosts viewed Bloom as relatively far upstream from competition among AI products, reasoning that data centers will need electricity regardless of which AI platforms prevail.

Takeaways

• The transcript offers a positive AI-power-demand thesis for Bloom, but also makes clear that the stock can swing sharply. The discussion did not give a price target or specific entry point.

Meta Platforms (META)

• Meta announced a competitor to Reddit during the show, prompting a short-lived reaction in Reddit shares.
• The hosts were skeptical that Meta’s new forum would rival Reddit, citing differences between Meta’s personal-relationship-based platforms and Reddit’s interest-based communities. They also pointed to Threads as an example of Meta’s difficulty expanding into this type of discussion.
• Meta’s large social platforms and AI efforts were discussed, but the show did not present a broader investment thesis or a specific recommendation on Meta shares.

Takeaways

• The Reddit product announcement is a potential competitive headline to monitor, but the hosts viewed it as a limited threat rather than evidence that Reddit’s position is already undermined.

Shopify (SHOP)

• The hosts described Shopify as an apparent potential beneficiary of agentic shopping, in which AI agents help consumers find and buy products.
• They also raised an open question: if AI companies build shopping and transaction tools directly into their agents, they might have less need to rely on Shopify or similar platforms.

Takeaways

• The transcript highlights both potential upside from AI-assisted commerce and a meaningful uncertainty about whether AI agents will strengthen or bypass Shopify’s role. It offered no firm conclusion or recommendation.

Yelp (YELP)

• The hosts discussed Yelp as a possible short-term sympathy trade if Reddit announced an acquisition or a major data-licensing deal that lifted Reddit shares.
• One host said he dislikes Yelp and considers its reviews less reliable than Google reviews. The idea was explicitly framed as a brief, conditional trade—not as a long-term investment thesis.

Takeaways

• This was a highly speculative trade idea tied to a hypothetical Reddit catalyst. The discussion did not express confidence in Yelp as a standalone investment.

Shutterstock (SSTK)

• Shutterstock was mentioned alongside Yelp as another company whose shares might move in sympathy with Reddit if Reddit announced a major acquisition or data-licensing deal.
• The connection drawn was that Shutterstock owns a library of visual content that could be relevant to AI data demand. The hosts described the potential trade as a brief arbitrage window, not a long-term thesis.

Takeaways

• The proposed opportunity depends on a hypothetical Reddit announcement and a short-lived market reaction. No price target, timing, or standalone bullish view on Shutterstock was provided.

Alphabet (GOOGL/GOOG)

• The hosts said Google has a potential data advantage through services including YouTube and possibly Gmail, while noting that the rights to use Gmail data for AI training were unclear.
• They described YouTube as a source of varied, original content, but also noted that AI-generated videos are increasingly appearing on the platform.
• The hosts mentioned rumors that an upcoming Gemini release could be especially strong, while acknowledging that this was unconfirmed.

Takeaways

• The discussion points to Google’s owned platforms as a possible data advantage, but also highlights uncertainty about data rights and the growing presence of AI-generated content. Treat the Gemini comments as rumor, not confirmed product evidence.

Apple (AAPL)

• One host said he would prefer Apple to win in personal AI assistants because of the prospect of processing information on-device or through private cloud computing.
• The group discussed a future Siri experience that could understand what is on a user’s screen and perform tasks such as adding items to a calendar. No investment case for Apple’s shares was developed.

Takeaways

• Privacy and on-device processing were presented as potential differentiators for Apple’s assistant products. The transcript does not establish how successful those products will be or offer a stock recommendation.

NVIDIA (NVDA)

• NVIDIA came up in a discussion of AI infrastructure spending. A host said Amazon’s internal chip work could make it less dependent on purchases from NVIDIA and other chip suppliers.

Takeaways

• The mention was limited to Amazon’s potential to rely more on its own chips. The transcript did not offer a direct bullish or bearish view on NVIDIA.

AI Data and Licensing

• The central investment theme was a possible shortage of fresh, high-signal human data as AI models increasingly train on material generated by other AI systems.
• Reddit, X, YouTube, and TikTok were cited as sources of current human discussion or content. The hosts suggested that AI labs may need to license access to valuable data, particularly for timely information and human judgment.
• They also noted alternatives such as older books, but argued that these do not provide the same up-to-date discussions of current events, products, and experiences.
• The thesis remains uncertain: the hosts said they did not know how much AI-generated content was already entering online platforms or how the issue would develop.

Takeaways

• The discussion suggests monitoring companies that control distinctive, actively updated data and have the ability to charge commercial users for access.
• The opportunity depends on data remaining genuinely human, useful for training, and legally available for commercial licensing. Those assumptions were not resolved in the episode.

AI Infrastructure and Market Risk

• The hosts argued that agentic AI could increase computing demand because agents may run many tasks simultaneously. They also described rising compute prices as a positive for infrastructure providers.
• At the same time, they emphasized uncertainty about how quickly AI will change advertising, shopping, education, and business operations.
• Specific risks raised included high market debt, concern about interest rates, rapid changes in AI capabilities, possible safety and regulatory issues, and unpredictable stock reactions to new information.

Takeaways

• The transcript is broadly constructive on long-term AI infrastructure demand but cautious about near-term volatility and which companies will ultimately capture the value.
• Chris explicitly advised investors to be careful with excessive leverage and not to become overconfident about market moves over days or weeks.

Anthropic (Private) and OpenAI (Private)

• The episode said Claude was conducting 26% of Anthropic’s own research, up from less than 1% in February, and that Anthropic had 30,000 AI agents running experiments, writing code, and helping build future models.
• The hosts discussed the possibility of an Anthropic IPO and mentioned a hypothetical valuation near $2 trillion, while Chris said that claims circulating online about the company’s finances and computing commitments were being misinterpreted.
• OpenAI was mentioned in connection with a delayed model release and the challenge of licensing human data. Neither company is a publicly traded stock.

Takeaways

• The discussion portrays AI labs as both rapidly advancing and potentially important buyers of data and computing capacity. Any IPO valuation or claims about company finances should be treated cautiously; the episode did not provide a confirmed offering timetable or investment recommendation.