Subsea Cables & Internet Infrastructure

Stocks of publicly traded companies including Coreweave, META, Microsoft, and Oracle have lost 4% to 39% of their stock market value over the last year. Coreweave's core business is building data centers stocked with GPUs to rent AI service providers for estimation and inference. Obviously, this makes the standard data center look downright capital light. Racks must be populated with servers and GPUs and capable of handling 120Kw power loads. So power infrastructure and backup alone cost many multiples more on a per square meter basis than the standard telecom hotel. Moreover, these data centers must be bigger because a large language model might have trillions of parameters to be estimated. Coreweave's massive debt loads have led to a 37% decline in its market value over the last twelve months.  Secondly, customer switching costs are very low in the AI data center market. AI service providers are software companies. They download their data into the bare metal ...

Zuckerberg should pull his head out of his ass. Yann LeCun's META departure underscores just how much Mark Zuckerberg has pivoted away from seeking AI innovation to chasing the latest fad, namely large language models like ChatGPT. Zuckerberg has a history of bad decisions. The stillborn Metaverse is just one example. Yann LeCun is one of the AI Godfathers. He used convolutional neural nets to greatly improve optical character recognition and computer vision. In contrast, large language models are an obvious dead end due to their creative writing tendencies and amoeba-like reasoning abilities (no offense, amoebas, you actaully excel ChatGPT). Fundamental breakthroughs are not achieved by scaling up digital parrots. Yann has repeatedly noted that the best AI models, the large language models, appear to be just regurgitation machines without any ability to reach new conclusions. They cannot identify or correct their own mistakes or realize that their approach is failing and adopt a n...