Wall Street Millennial

Huang uses a very manipulative rhetorical technique. He has no evidence that token maxing produces a return on investment. So rather than defend that claim directly, he leans on a larger unstated premise that AI dramatically boosts engineering productivity.

— Wall Street Millennial, at 19:05

Overall, it's a good thing that we're finally starting to see some real alternatives to Nvidia. More competition should eat away at Nvidia's monopoly pricing power and eventually bring down the cost of AI compute for everyone.

— Wall Street Millennial, at 16:38

The likes of Nvidia and Microsoft need to keep OpenAI alive so as to keep the AI bubble going and support their own stock market valuations. That's why they're participating in OpenAI's funding round. But I don't think Jensen Huang actually believes that OpenAI will cure cancer o

— Wall Street Millennial

While that's not a huge growth rate, it's not bad when you consider how large and mature the company already is. There is zero evidence that enterprises are vibe coding their own CRM solutions and canceling their Salesforce subscriptions... Customers are probably waiting for the

— Wall Street Millennial

Mark Beni off made huge claims about AI which generated a lot of media headlines, but there's really no evidence that Salesforce is getting any efficiency gains from AI.

Microsoft says that AI usage is no longer optional and is now incorporated into performance reviews... An anonymous Microsoft employee told the pragmatic engineer that he has resorted to unproductive token maxing.

— Wall Street Millennial, at 9:22

The shift to AI coding has coincided with huge problems with Windows 11, which is plagued by bugs and security issues. Despite bold claims about AI replacing humans, Microsoft's number of full-time employees has continued to increase.

Alphabet understands that the surging demand for AI tokens is probably not sustainable. But if that's the case, why are they still spending so much on AI data centers? ... I think at some level Sundar Pachai and Alphabet senior leadership do understand that the current AI spendin

— Wall Street Millennial, at 13:33

Of the big three logic chip manufacturers, Intel is the least competent. As we discussed earlier, they cannot manufacture the highest end chips and are forced to outsource this to TSMC. So, how is Intel supposed to teach Tesla how to do something which Intel itself cannot do?

— Wall Street Millennial

It's interesting that the leaderboard is called Claudonomics. Meta has spent billions, if not tens of billions of dollars, to build its own llama family of AI models. But Llama is so bad that not even Meta's own employees use it.

— Wall Street Millennial, at 12:31

The entire idea of a Bitcoin treasury company is pretty dumb... Now that the share price premium has evaporated, he's desperately trying to justify his company's continued existence by pretending he has a sophisticated strategy to somehow create value. In reality, all he's doing

— Wall Street Millennial

What makes Karp's narrative so compelling is that there is some truth behind it. Palanteer's business is indeed doing very well and they are one of the few large software companies to successfully integrate AI. Under normal circumstances, such a company would already have a rich

— Wall Street Millennial

Musk doesn't care about Tesla's minority shareholders... Musk has previously diverted engineers and NVIDIA GPU orders from Tesla to X and XAI.

— Wall Street Millennial