1. $NVIDIA(NVDA)$ Why is NVDA down today? 1. Revenue growth slows as dominance in pre-training turns to competition in inference. 2. Margins fall because of ⬆️ 3. Multiples compress from sky-high levels to “normal”. The stock could fall 80-90% depending on the change in growth and margins.90% of beating the market is really just:Buy great companiesDoing nothing for YEARSProfitIt's harder than it looks, but the formula is simple.2. $SoFi Technologies Inc.(SOFI)$ SoFi stock now only up 40% in the past 3 months. What a disaster.ImageResults look great. The reaction may be negative because it wasn’t a blowout. But long-term a 24% revenue growth rate is phenomenal and more people are CHOOSING to use the produc
1.Buy & Hold?Hold on a minute...If you buy and hold the S&P500 $.SPX(.SPX)$ index, you're not actually buying and holding, you're really buying an interest in a constantly shuffling set of stocks -- and in 10 year's time you may well not recognize 1/3 stocks in your portfolio!Image2.Non-rec Bulls 🤔 This chart shows the history of Bull markets coming out of Recession (most often large, triple digit gains) vs those *not* preceded by recession circled in blue (still significant but shorter, smaller by comparison). One implication is that the current one looks about fully-cooked.3."Cyclically-Adjusted Earnings" This chart tells you all you need to know about US vs global equity relative performance...(and likely will hold key clues in terms of
The debates around DeepSeek are intense - US vs. China, big vs. small models, open vs. closed source, and the shockingly efficient architecture it represents. Pride, fear, disbelief, disgust - all these emotions have clouded the facts. A few personal thoughts:Thoughts on Training Costs: 1⃣ $6M Training Costs = Plausible IMOQuick math: Training costs ∝ (active params * tokens). DeepSeek v3 (37B params; 14.8T tokens) vs. Llama3.1 (405B params; 15T tokens) = v3 theoretically should be 9% of Llama3.1's cost. And the disclosed actual figures aligned with this back-of-the-envelope math, meaning, the number are directionally believable.ImagePlus, there was no hiding, the footnote clearly said: “the aforementioned costs include only the official training of DeepSeek-V3, excluding the costs associa
Filter bubble burst leading to revaluation - an excellent buying opportunity
Pure panic selling and deleveraging market conditions - time to buy the dip. We might see several such opportunities this year. $NVIDIA(NVDA)$ It's now confirmed that this major drop was mainly triggered by DeepSeek. Why did DeepSeek, released earlier, cause today's drop?Because today DeepSeek topped Apple's App Store free app download rankings in both China and US regions, surpassing ChatGPT in US downloads.DeepSeek R1's significantly reduced AI training and inference costs, and this top ranking symbolizes global recognition of this advantage, directly competing with and outperforming US AI.QiMai Data screenshotActually, DeepSeek's progress isn't negative for the AI industry; rather, it accelerates the entire sector. Industry upgrade and scale ex
NVIDIA -11%! Will DeepSeek Drag Computing Demand Down?
$NVIDIA(NVDA)$ opened 11% down, with the market dubbing DeepSeek "the biggest threat to the US stock market."What’s even more striking is the ultra-low cost of achieving such remarkable results, raising shareholder skepticism about the massive spending by other large models. DeepSeek-V3 trained a 671 billion-parameter model using only 2,048 H800 GPUs, costing just $5.576 million—significantly lower than the training costs of other top-tier models (e.g., GPT-4 at $1 billion).NVIDIA and other hardware manufacturers traditionally hold an advantage on the training side, but this development could potentially impact their market position and strategic direction. Some believe DeepSeek may disrupt NVIDIA’s dominance in the AI hardware space.Will computin
AI Fall Before MSFT Earning Release, Should You Buy or Sell?
$Microsoft(MSFT)$ Microsoft is set to report its quarterly financial results after the markets close on January 29, 2025, sparking a lot of investor interest in whether they should buy Microsoft stock before earnings. In this video, I'll answer that question and preview what investors should be watching for in the upcoming earnings release. So, let’s take a closer look at this intriguing AI powerhouse. Microsoft last updated investors on October 30, 2024, revealing a 16% increase in revenue to $65.6 billion. However, operating income only grew by 14% to $30.6 billion, which was a bit surprising. Typically, when Microsoft’s revenue rises by 16%, operating income sees a more significant boost due to the company’s strong operating leverage. Operating
$SPDR S&P Semiconductor ETF(XSD)$ I don't think the sell-off in semiconductor names really changes the AI narrative at all. AI training costs are falling precipitously - something like 90% a year - so it is only natural that companies can gradually produce equal or better AI models at cheaper prices. This time, there just happens to be a Chinese team who were very clever in doing it, but it is more of a sign to come - and something to just get used to - rather than a thesis-breaker. Today, the market is pricing in the cost decline, but not the explosive productivity gains that we will eventually see. On other occasions in the future, the opposite will occur. The volatility may be scary, but tech stocks are inherently volatile,
$ARM Holdings(ARM)$ ARM Holdings is a powerhouse in the semiconductor industry, and its pivotal role in shaping future technology cannot be understated. A key highlight of ARM's journey is its close collaboration with NVIDIA, a tech giant leading the AI revolution. While NVIDIA's attempted acquisition of ARM faced regulatory hurdles, their ongoing partnerships signal immense synergy in areas like AI, GPUs, and high-performance computing. ARM's importance is further amplified by global initiatives such as the Trump Stargate AI program, which emphasizes cutting-edge AI innovation and infrastructure development. ARM's architecture, being widely adopted in AI-driven devices, positions the company to benefit significantly from such initiatives. Add
$NVIDIA(NVDA)$ Let the selling begins. The Americans and the West have failed to realize that the Chinese are very good in copying and re-engineering. For decades, the Chinese are good copying because they do not care about Intellectual property rights. Money is the number one pushing factor, nothing else matters. It is stupid that the West suddenly realized it even in the AI space. Sanction chips and technologies have no impact on the Chinese because there are thousands of ways to get the parts. There is no way for the US to stop the Chinese.
These moves suggest optimism for the sector, especially as AI transforms energy management, optimizing grids and improving plant efficiency. Power companies embracing AI and renewable energy are likely to benefit from long-term growth. Notable stocks to watch include Vistra (VST), NextEra Energy (NEE), Duke Energy (DUK), and Xcel Energy (XEL), all well-positioned in the evolving energy landscape. While power stocks may not double immediately, the sector’s growth outlook remains strong. The shift toward cleaner energy, AI advancements, and potential political support make it a good time to invest, though risks like regulatory changes and energy price fluctuations should be considered.
Ok, folks. It seems we are in a mini crash, again. It is gonna get worse before it gets better, I assume. Big players try to accumulate at a lower price, while some weak hands try to get out when they can. Also, this deep seek story is a total psyop. Don't you feel it is quite odd for them to release this after a few days of DJT's big stargate AI initiative? [Happy]
China AI Just Won The Tech Race and It Change The World!
$Palantir Technologies Inc.(PLTR)$$NVIDIA(NVDA)$$Meta Platforms, Inc.(META)$$Tesla Motors(TSLA)$ What Is Deepseek R1 China has just introduced a groundbreaking new technology that’s shaking up the global tech scene. What’s even more intriguing, though, is the timing of its release and what it signals for the future of US-China relations. On January 20th—the very same day Donald Trump was inaugurated as the 47th president of the United States—China unveiled Deepseek R1, an advanced AI tool designed for deep data analysis. This new development positions itself as a direct competitor to America's OpenAI. While Google and
The Rise of Quantum Computing Stocks: Is Now the Time to Invest?
Quantum computing, once a futuristic dream, is rapidly evolving into a tangible technology poised to revolutionize industries like finance, healthcare, and cybersecurity. With major tech players and startups racing to achieve quantum supremacy, is this the golden era for quantum computing stocks? Key Updates Major Industry Developments IBM unveiled its 1,000-qubit quantum processor, "Condor," marking a significant leap in quantum processing power. Google( $Alphabet(GOOG)$ ) and Honeywell( $Honeywell(HON)$ ) have also made strides in quantum error correction, a critical factor for scalable quantum systems. Funding and Valuations Quantum startups raised over $3 billion globally in 2024, led by IonQ(
EV Battery Breakthrough: Can Solid-State Technology Charge Up Stocks?
Solid-state batteries are being hailed as the next revolution in electric vehicles (EVs). With higher energy density, faster charging times, and improved safety over traditional lithium-ion batteries, this innovation has attracted massive investments from automakers and tech giants. But can this new technology propel EV battery stocks to new heights? US firm's all-solid-state EV battery breakthrough offers more energy density, longer range Key Updates Solid-State Batteries: The Real Game Changer for Electric Vehicles? Major Innovations Toyota unveiled its prototype EV with a solid-state battery, boasting a range of 745 miles on a single charge and a 10-minute recharge time. QuantumScape (QS) announced a breakthrough in scaling solid-state battery production, targeting commercial availabil