As foundational AI models become increasingly similar, their core technology is at risk of becoming a commodity. The true long-term investment opportunity is shifting to companies that can build valuable applications and integrate AI into existing products. Focus on companies with massive distribution channels, as they are best positioned to capitalize on this trend. Key players to consider are Microsoft (MSFT), Google (GOOGL), and Amazon (AMZN), who back the major models and own dominant platforms. Investors should prioritize these large-cap tech companies that control the application and distribution layers of the AI ecosystem.

Artificial Intelligence (AI) & Large Language Models (LLMs)

• The discussion centers on the rapid evolution of Large Language Models (LLMs) like ChatGPT, Claude, and others.
• A key observation is that these models are being trained on each other's outputs and knowledge bases. This causes expertise from one model to "bleed" into others.
• An example was cited where multiple AI models were tasked to discuss and choose the best answer among themselves, and they often concluded that ChatGPT's answer was the best one.
• The speaker suggests that this cross-pollination of knowledge will eventually lead to the foundational layer of all major AI models becoming very similar or homogenous.
• This implies that even if a company adds new, special data to its model, that knowledge can quickly spread to competing models simply by having the AIs interact with each other.

Takeaways

• The AI Arms Race: The conversation reinforces that AI is a critical and competitive field. The companies behind the major models mentioned—OpenAI (ChatGPT), which is heavily backed by Microsoft, and Anthropic (Claude), which is backed by Google and Amazon—are the key players to watch.
• Focus on Application, Not Just the Model: The core insight is that if foundational AI models become a commodity, the long-term competitive advantage may not come from having the single "best" model.
• Investors should consider which companies are best positioned to build valuable applications and user experiences on top of these powerful models.
• Distribution is Key: The value may lie with companies that have massive distribution channels to integrate AI into products people already use. The mention of Bing (Microsoft's search engine) is an example of this integration strategy.
• Potential for Commoditization: The homogenization thesis suggests that the unique technological edge of any single foundational model could decrease over time. Investors should be aware that the primary value may shift from the core technology to the data, application, and distribution layers.