Sebastian GdeV

[Anthony Pompliano 🌪 @APompliano]

Why Most People Are Wrong About Deepseek AI and The Impact on Financial Markets

A new artificial intelligence model out of China is creating uncertainty and volatility in the market this morning. Deepseek, which was supposedly developed by a startup with ~ 200 employees, has proven to be more effective than OpenAI’s ChatGPT in a number of tests.

The mobile app associated with Deepseek is currently sitting at #1 in the charts in the United States and China. Usually a launch that gains momentum like this so quickly is celebrated, so why is the market freaking out?

The short answer is because majority of investors don’t understand how technology and financial markets are related.

In order to understand the miscalculation, we must first understand what the average investor believes right now. They have seen a new Chinese model published that was reportedly created for under $6 million and the LLM has been open-sourced for anyone to use. This immediately begs the question of why American companies have spent so much money to build inferior models, along with creating a narrative that the US may be losing the competition for AI supremacy.

But here is the thing — you can’t believe anything coming out of China right now. The government has a track record of misleading financial markets on various metrics and Chinese companies have been caught making fraudulent claims over and over again.

I have no idea how much money it cost to build Deepseek. But it feels odd that OpenAI, Anthropic, Perplexity, xAI, and many others have raised billions of dollars each and then a random startup claims they built something better for less than 1% of the capital investment. Could it be possible? Sure. Is it probable? Absolutely not.

But lets assume that Deepseek actually pulled this off for $6 million. The next question is whether investors should be excited or scared — I feel strongly that a better, cheaper model available to anyone in the world should be a net positive for society. You are going to hear everyone talking about Jevan’s paradox in the coming days. The idea states an increase in efficiency in resource use will generate an increase in resource consumption rather than a decrease. (Even Microsoft’s Satya Nadella tweeted it already!)

More companies are able to leverage the technology to create economic activity and drive GDP growth. Individuals should be able to save time and become more effective at their jobs. And more problems will be solved. So Deepseek’s innovation will drive more use of AI, rather than less.

This is all positive on a macro basis.

The fear starts to come into investors’ minds when they wonder if they have bet on the wrong companies. Many of the investors I know couldn’t explain the intricacies of artificial intelligence. They throw around buzz words and hope no one digs below surface level in their conversation. Instead of allocating capital based on technical understanding, most of these individuals and institutions are merely momentum investing.

NVIDIA has gone up a lot, so they buy the stock because odds are it will continue going up. Same with a variety of other companies. The funny thing is that momentum investing in growth stocks is a good strategy most of the time. However, it can be brutal to go through the few times that momentum investing is the wrong strategy.

So ask yourself — why are investors selling NVIDIA because a better model came out? Shouldn’t NVIDIA investors be excited that AI will become more prevalent and NVIDIA’s products will be used more often?

The concern this morning is Deepseek claims they built the new model using inferior chips to what many American companies have access to. This is supposed to make people bearish on NVIDIA, but instead I think people should be more bullish since NVIDIA’s competitive advantage is so obvious. For example, if Deepseek could do this with inferior chips, imagine what they could do with NVIDIA products?

This brings me to how I see the market today. First, AI models are largely thought to be infrastructure but I don’t agree with that framing. I see them more as cars. All cars have four wheels and an engine, but they come in a variety of wrappers. Some people like fast cars and some like big cars. Some people like red cars and others like black. Some people like Ferrari and others like Tesla. There is something different for everyone.

Models are similar in that it will not be a winner-take-all market. There will be different models for different use cases. It makes no sense to invest capital in a single model hoping it is the one model to rule them all. That is not how technology works.

Next, I am a firm believer that investing in companies that USE artificial intelligence, rather than MAKE artificial intelligence will be the right way to play this market. There may be one or two model producers that accrue significant value, but I am not trying to pick the one needle in a haystack. I simply want to have above average returns across the sector. As Howard Marks points out, if you try to be the top performer every year, then you have to be willing to be the bottom performer when you are wrong.

I am spending a lot of time looking for companies that are using AI to drive down expenses and increase productivity. One metric that I continue to evaluate is revenue per employee. I don’t want to merely see a static measurement, but instead I want to see this number improving over time. If a company starts with $500,000 of revenue per employee and two years later it has $1.2 million in revenue per employee, this is a company that I would be very interested in understanding better.

Artificial intelligence should make a company more efficient and more productive. This metric is a quick way to quantify whether a company understands that benefit or not.

Lastly, I am paying attention to the intersection of AI and crypto. There are many ways to play the intersection, but the area I am more interested in is the monetization of open-source technology. I have a small position in the ai16z token, which is a crypto coin related to the popular Eliza framework, because I believe there is immense value to be created and captured by open-source teams if they can figure out how to create open-source technology with economic incentives attached to the project. This can not only help attract capital for future development, but you can create an entirely new incentive system to attract intellectual capital to help push a project forward.

There will be other opportunities at this intersection, including AI hedge funds, stablecoin payments, and AI workers, but the monetization of open-source technology feels like one of the largest opportunities.

This brings me back to where we started — the market is melting down today. NVIDIA was down more than 11% this morning and stocks across the board were crashing. Every mainstream media outlet was talking like financial markets were under attack. Even bitcoin fell below $100,000 in the early morning hours (the horror!).

But I think you should take a deep breath and see this situation for what it is. The market is testing the emotional discipline of investors. As Buffett always says, the market is a voting machine in the short-term, but a weighing machine in the long-term. Don’t fail the test.

I believe the best way to invest at the edges of innovation is to apply timeless investing principles to new innovations. You want to look for compounding businesses, even if they aren’t necessarily throwing off cash today, and optimize for the companies that will have the resilience to thrive regardless of market conditions.

We are watching a multi-decade situation start to play out. It is early innings. You can put capital into artificial intelligence companies. You can put money into bitcoin and crypto. And you can put money into companies using AI and bitcoin. There is opportunity everywhere.

The best time to be alive will be tomorrow. The second best time is today. We are watching an explosion of economic activity and I predict it will continue for the next few years. There will be volatility along the way, but you need volatility if you want appreciation.

Be thankful we aren’t living through a dead money period. Keep your head on straight and don’t panic buy or sell. Everything you want is at the end of your long-term thinking process. You just have to survive to get there.

[Sebastian GdeV @sebastiangvm96]

@Wealthsimple #WealthsimpleCrypto #contest

[Wealthsimple @Wealthsimple]

BTC is still over $100,000 CAD. 🇨🇦

[Walker⚡️ @WalkerAmerica]

“Why do these silly #Bitcoin maximalists keep saying the gov/Fed prints money??”

[Sebastian GdeV @sebastiangvm96]

@rovercrc Wait, moonboy-rover doesn’t think up only? 🤔

[Sebastian GdeV @sebastiangvm96]

@rovercrc Good until you woke up!

[Sebastian GdeV @sebastiangvm96]

@dehartigewortel @TNTcrypto007 @rovercrc Ay this point he’s gave away all his BTC for how many times it’s dipped below 20k.

[Sebastian GdeV @sebastiangvm96]

@rovercrc You hype bitcoin high, you hype bitcoin low.
You hype bitcoin everyday, making you extremely permabull!

And… All for the follows.

[Sebastian GdeV @sebastiangvm96]

@rovercrc Of course we will dipshit, we’ll brag of our returns because we bought at $20 k. Saying “and we bought at $20k can you imagine?!”

[Sebastian GdeV @sebastiangvm96]

@rovercrc Dude you’re all hype, no substance. You do this for the follows. I will LMFAO when this fails.

[Sebastian GdeV @sebastiangvm96]

@rovercrc Dude you’re all hype, no substance. Geeez. You do this for the follows.

[Sebastian GdeV @sebastiangvm96]

@rovercrc Dude you’re all hype, no substance. You do this for the follows.

[Sebastian GdeV @sebastiangvm96]

@rovercrc Dude you’re all hype, no substance. You do this for the follows.

[Sebastian GdeV @sebastiangvm96]

@rovercrc Dude you’re all hype, no substance. You do this for the follows.

[Sebastian GdeV @sebastiangvm96]

@mikealfred @CoinDesk So Celsius files for bankruptcy, and Goldman purchases the assets fire sale from the pool of depositors.
Against the depositors wishes to sell their assets.
To finance operations, depositors get A fraction of what they owned. If any?

[Sebastian GdeV @sebastiangvm96]

@mikealfred @CoinDesk Why does Goldman want to buy out Celsius? In your opinion, is the business model profitable but bad management? What does it mean for the cel users?

[Joshua Browder @Joshuabrowder]

If you have under $10,000-25,000 with @CelsiusNetwork (or any crypto exchange that has frozen your funds), here is a step by step guide on how to get your money back ahead of everyone else using the law:

[Sebastian GdeV @sebastiangvm96]

@otisa502 @CelsiusNetwork @Nicholas_Merten @SimonDixonTwitt @100trillionUSD @NewsAsset

[Otis @otisa502]

1/16 Let’s wait to hear an official statement from @CelsiusNetwork. But if that statement is positive, then this is GameStop, AMC and Wall Street Bets all over again (the sequel.)

These are my thoughts on the short sellers. I can only speak for that part of the attack on

[Money Lisa @real_moneylisa]

I found an old $ETH wallet with 10 ETH in it!! Because old me was smart, I'm giving away 2 ETH to someone who follows me, likes and retweets this. 24 hours - go!