Fri, 23 Feb 2024 19:48:00 +0000
Nvidia's revenue will continue to grow at large scale (i.e., remain at or above its early-2024 run-rate and not materially contract) for approximately the next 2–3 years from February 2024.
“I think the revenue scale will continue for like the next 2 or 3 years probably for Nvidia.” View on YouTube
Nvidia's revenue continued to grow at a large scale through 2025-2026, with data center revenue reaching well over $190 billion, far exceeding its early-2024 run rate.
Consensus Wall Street analyst expectations as of February 2024 are that Nvidia will still hold roughly "60-something percent" of the GPU market used for AI workloads five years later (around 2029).
“The Wall Street analysts I've been listening to think that in five years they're still going to have 60 something percent market share, so they're going to have a substantial percentage of this market or this Tam.” View on YouTube
Nvidia's AI accelerator market share remained dominant (roughly 75-85%) through 2026, tracking toward but modestly below the '60-something percent by 2029' forecast being discussed, given it started from a much higher base than expected to decline that fast.
The AI-driven wave of new applications (both consumer and enterprise) powered by cloud data centers and GPUs will continue to expand for at least the next decade from 2024, i.e., through at least 2034.
“I mean, we're just at the beginning of a wave that's probably going to last at least a decade.” View on YouTube
The AI application wave continued to expand rapidly through 2025-2026, well on pace for at least a decade-long growth cycle as predicted.
If Groq ("grok with a Q") is able to produce its LPU chips at scale, its market capitalization will increase substantially from its early-2024 level of roughly "a billion something" (i.e., it will become worth significantly more than a low-single-digit billion dollars over the following years).
“there's a lot of market cap for grok to gain by just being able to produce these things at scale.” View on YouTube
Groq's valuation grew dramatically from roughly a billion dollars in early 2024 to a reported valuation in the billions (with a major 2025 funding round reportedly valuing it around $6.9 billion), confirming substantial market cap growth.
Following the public backlash to Gemini’s rollout (Feb 2024), Google will not remove ideological bias from the model but will instead adjust it to be less obvious and more subtle, rather than changing the underlying values or objectives.
“Now, I think what's going to happen now is in light of this, the reaction to the rollout is, do I think they're going to get rid of bias? No, they're going to make it more subtle... I think they're simply going to they're going to dial down the bias to be less obvious.” View on YouTube
Google did adjust Gemini's behavior following the February 2024 backlash, but the company also made more overt public statements acknowledging and correcting specific failures rather than purely making bias 'more subtle.'
General-purpose AI Q&A services (such as Google Gemini) will only be commercially successful if they implement user-level personalization of values and filtering; if they do not, over time a significant share of users will abandon these products after encountering outputs that conflict with their objectives.
“I think the only way that a company like Google or others that are trying to create a general purpose, knowledge Q&A type service are going to be successful is if they enable some degree of personalization... If they don't allow this, eventually everyone will come across some search result or some output that they will say doesn't meet their objectives... they're going to say, I don't want to use this product anymore.” View on YouTube
AI chatbot personalization features did expand across providers by 2025-2026 (custom instructions, memory features), partially matching the prediction, though it is difficult to isolate personalization as the sole determinant of commercial success.
Due to dissatisfaction with heavily filtered/bias-tuned large models, multiple competing AI models will proliferate and open‑source models will gain a significant competitive advantage ('win') in the market over more tightly controlled proprietary models.
“And so it is actually an opportunity for many models to proliferate, for open source, to win.” View on YouTube
Open-source AI models (Llama, DeepSeek, Qwen, and others) proliferated significantly and gained substantial market and mindshare through 2025-2026, consistent with the prediction.
In response to criticism of Gemini, Google will not materially change the ideological framing of the model but will instead make the biased behaviors less overt and harder to detect, rather than eliminating them.
“what I'm afraid of is that what Google will do is not change the underlying ideology that this AI model has been trained with, but rather they'll dial it down to the point where they're harder to call out. And so the energy will just be more subtle.” View on YouTube
Google continued to iterate on Gemini's behavior and safety tuning through 2025-2026 without a clear public acknowledgment of an underlying ideological framing being deliberately obscured rather than corrected.
If Google were to massively license high‑quality training data (on the order of tens of billions of dollars per year) and position itself as the most reliable, truth‑focused AI provider, this strategy could support Google achieving a market valuation on the order of $10 trillion in the future.
“We are going to be the truth tellers in this new world of AI. So when everybody else hallucinates, you can trust Google. To tell you the truth, that is a $10 trillion company, right?” View on YouTube
Google did not reach a $10 trillion market valuation as of mid-2026; its market cap remained in the $2-3 trillion range, well short of the predicted figure.
Within a few months after February 2024, the quality of LLMs will improve so quickly that people will look back on early‑2024 LLM behavior (and this discussion) as primitive and notably bad, analogous to how 1996 web search compares to later eras.
“I think we're like in the 1996 era of Llms, and in a couple of months, the pace, things are changing. I think we're all going to kind of be looking at these days and looking at these pods and being like, man, remember how crazy those things were at the beginning and how bad they were?” View on YouTube
LLM quality improved dramatically within months of February 2024 (GPT-4o, Claude 3, and later models), and early-2024 model behavior is now widely viewed retrospectively as primitive compared to 2025-2026 systems.
If Google continues on its current Gemini/product-ideology path, it will lose a meaningful number of users/consumers to competing AI/search products that offer a better experience; competitive market forces will correct for Google’s missteps rather than allowing it to retain dominance with an inferior product.
“I actually think to your point, Google could be going down the wrong path here in a way that they will lose users and lose consumers, and someone else will be there eagerly to sweep up with a better product. I don't think that the market is going to fail us on this one” View on YouTube
Google's Gemini did face significant early competitive criticism, but the company mounted a substantial comeback in AI competitiveness by 2025 with strong Gemini model releases, complicating a clean narrative of losing users to competitors.
No single company will be able to achieve a monopoly on training data or information control for AI models through paid content licensing, because the open internet already contains sufficient data to prevent effective monopolization and a resulting 'disinformation age' driven by one actor.
“I think the open internet has enough data that there isn't going to be a monopoly on information by someone spending money for content from third parties. I think that there's enough in the open internet to give our all. Give us all kind of, you know, the security that we're not going to be monopolized away into some disinformation age.” View on YouTube
No single company achieved a monopoly on AI training data or information control through paid licensing deals as of 2026; the open internet and open-source model ecosystem continued to provide broad competitive access to training data.
If Transnistria (a Russian enclave in Moldova) formally seeks annexation by Russia and/or holds a referendum to join Russia following the February 2024 discussions, Western governments will treat this as evidence of broader Russian expansionist aims and the Ukraine conflict will significantly escalate as a result.
“So now Transnistria is right there and could theoretically make a play to try and join Russia. Why do I think this is a big deal? Because if something like this happens, it could really expand the Ukraine war. The West is going to use this as evidence that Putin wants to invade multiple countries and invade, you know, a bunch of countries in Europe. And this could lead to a major escalation in the war.” View on YouTube
Transnistria did not formally seek annexation by Russia or hold a referendum to join Russia following this discussion, and no major escalation of the broader Ukraine war stemmed from Transnistria specifically.