Is DeepSeek inaccurate? … claims of plagiarism … how DeepSeek’s technology works … the risks it represents … a roundtable discussion with our experts The DeepSeek story continues to evolve with new information emerging. Let’s begin with Reuters: Chinese AI startup DeepSeek's chatbot achieved only 17% accuracy in delivering news and information in a NewsGuard audit that ranked it tenth out of eleven in a comparison with its Western competitors including OpenAI's ChatGPT and Google Gemini. The chatbot repeated false claims 30% of the time and gave vague or not useful answers 53% of the time in response to news-related prompts, resulting in an 83% fail rate, according to a report published by trustworthiness rating service NewsGuard on Wednesday. Meanwhile, as to how it caught up with the U.S. incumbent AI platforms so quickly, the answer might be “stealing.” Here’s The New York Post: OpenAI, the company behind ChatGPT, says it has proof that the Chinese start-up DeepSeek used its technology to create a competing artificial intelligence model — fueling concerns about intellectual property theft in the fast-growing industry. OpenAI believes DeepSeek, which was founded by math whiz Liang Wenfeng, used a process called “distillation,” which helps make smaller AI models perform better by learning from larger ones. While this is common in AI development, OpenAI says DeepSeek may have broken its rules by using the technique to create its own AI system… Security researchers at Microsoft, which has poured billions into OpenAI, discovered last fall that individuals with possible links to DeepSeek were harvesting vast troves of data through OpenAI’s application programming interface, or API, sources told Bloomberg. Despite this, Microsoft has been gracious and supportive of DeepSeek. From Bloomberg: “DeepSeek has had some real innovations,” [said Microsoft Corp. Chief Executive Officer Satya Nadella] during an investor call after Microsoft reported quarterly results on Wednesday. “Obviously now that all gets commoditized and it’s going to get broadly used.” Stepping back, let’s not rush to conclusions about what DeepSeek’s technology means for U.S. AI sector. New information and perspectives are emerging fast, and the story will continue evolving. So, for now, let’s look at what’s most likely accurate about DeepSeek, and analyze how it could influence your portfolio. By the way, our experts Louis Navellier, Eric Fry, and Luke Lango just sat down with our Editor-in-Chief and fellow Digest writer, Luis Hernandez to film a roundtable discussion on the topic. It’s your best way to get our experts’ unfiltered thoughts about DeepSeek, its impact on the AI sector, and what to do about it in your portfolio. More on that shortly… | Recommended Link | | | | Investing legend Louis Navellier warned us about the stock market crash of 1987... the 2000 dot-com crash… Enron’s collapse… and the 2008 financial crisis crash. He also predicted the rise of a host of iconic stocks… including Google, Apple, Amazon, Netflix, Facebook, and Nvidia. Today, he’s stepping forward to make history yet again… with a critical market forecast he’s calling: “My biggest prediction in 47 years…” Click here to see it. |  | | First, as the dust begins to settle, DeepSeek is appearing to be a major boon for the U.S. AI sector rather than a death knell However, its arrival does mean that the AI leaderboard is shifting slightly. There are some corners of the AI sector that could face headwinds, even if DeepSeek doesn’t live up to all its claims. To begin our analysis, it’s critical to understand how DeepSeek competes with incumbent U.S. AI models – despite operating at vastly lower cost using greatly reduced energy. The answer has to do with something called Mixture-of-Experts (MoE) architecture, which our technology expert Luke Lango detailed in yesterday’s issue of Hypergrowth Investing: Ever since the AI Boom began, the U.S. has consistently enforced export bans on AI chips to China, thereby limiting the number of chips Chinese firms can buy. As a result, Chinese developers were forced to embrace a “do more with less” mentality. At the heart of DeepSeek’s breakthrough is something called a Mixture-of-Experts (MoE) architecture. It is a model composed of multiple topic-specific sub-models. Therefore, when you ask DeepSeek a question, the only part of the model that “wakes up” is the expert sub-model relevant to your question. Thanks to this modular approach, DeepSeek can save an immense amount of computing power on each query because only part of the model is roused per query. This drastic reduction in activated parameters is partially what has allowed the firm to create an AI as good as leading models for ~95% lower costs. Now, lower costs would seem to be an enormous benefit for AI companies. So, why did so many AI stocks implode earlier this week after DeepSeek emerged? The fear behind a lower-cost AI buildout The selloff earlier this week reduces to a core concern… AI at lower cost/reduced energy consumption means that the AI Big Dogs (think the Mag 7) won’t need to spend billions on their AI development. And that would mean the chipmakers, datacenter operators, and component suppliers that were anticipating billions in tech development revenue will need to rethink their expectations. We’ll address this concern in a moment. But first, let’s not miss the obvious… Lower cost/reduced energy AI technologies will be a tremendous benefit for what Luke has called the “AI Appliers” – meaning the companies that use AI in their existing operations to improve their profit margins. For them, AI at vastly reduced cost means they can create and offer far more advanced versions of their products and services at greatly reduced expense. That’s good for their bottom lines. But they’re not the only ones benefiting. With an entire new cohort of businesses able to implement AI at lower costs, the companies that make AI applications possible could enjoy an avalanche of new revenue streams. Back to Luke: This DeepSeek breakthrough looks like a big-time opportunity. That’s why we’re telling our subscribers to look for plays in the “AI Application Layer,” with the companies developing cutting-edge AI applications… We would get exceptionally bullish on AI software and services stocks like Meta and Apple (AAPL) or companies like Spotify (SPOT), Intuit (INTU), ServiceNow (NOW) and Atlassian (TEAM) – those creating and deploying AI models and solutions. | Recommended Link | | | | I work at InvestorPlace, and while I’m NOT an investing expert... I feel like one when I read Eric Fry. He recommended a trade on Freeport McMoran that turned my $662 investment into more than $10,000 — one of the biggest wins of my life.* His unique trading strategy doesn’t find wins this big on a regular basis, but it has delivered dozens of gains of 100% and 200%. He just released a new video about this strategy and it’s “must-viewing” for anyone who wants the chance at big wins like the one I experienced Go here to check it out now… *The investment results described in this testimonial is not typical; investing in securities carries a high degree of risk; you may lose some or all of the investment. |  | | But what about the AI infrastructure companies that were expecting those billions in revenues from Big Tech? Let’s return to the potential losers of AI technologies at lower cost/reduced energy consumption. At first blush, high-end chipmakers and datacenter/power facilitators are in the crosshairs. After all, if AI technologies no longer require expensive, high-end microchips, that would hurt the bottom lines of companies such as Nvidia. Similarly, if AI technologies no longer require enormous volumes of energy, that would weigh on datacenter/power companies and their related component businesses. Is it time to bail on these stocks? Not necessarily. To make the case for why DeepSeek won’t be a death knell for AI infrastructure plays, let’s go to Christophe Fouquet, the CEO of leading chipmaker ASML. From his interview yesterday with CNBC: A lower cost of AI could mean more applications. More applications means more demand over time. We see that as an opportunity for more chips demand… For AI to really come to life in the next few years — not only with the hyperscalers [like Microsoft, Amazon and Google], but with all of us in our phone, PC — we need AI to address two things: cost and energy consumption. We believe that anything that will go in the direction of lowering cost on AI is, in fact, probably news because this will allow applications to go to many, many more devices. Okay, but what about AI datacenters? If DeepSeek’s technology operates at just a fraction of the energy used by U.S. incumbent platforms, that will mean far less power consumption than previously forecasted. This is why in a note to clients on Monday, Bank of America analysts flagged this concern, writing that DeepSeek is “raising doubts about the high expectations for…power requirements.” While this would appear to be a major stumbling block for AI power companies, “Jevons Paradox” suggests it’s the opposite. Back to Luke: Originated by 19th century British economist William Stanley Jevons, the Jevons Paradox states that improvements in a resource’s efficiency tend to increase – rather than decrease – the overall consumption of that resource. That’s because greater efficiency lowers a resource’s cost, which can lead to increased demand. In other words, DeepSeek’s technology could mean that the cost-per-unit of energy needed to run AI datacenters will drop substantially…but rather than hurt overall power consumption, this would lead to a tsunami of demand from new companies wanting AI at lower energy cost. So, though the cost of energy-per-unit could fall, overall energy demand could grow dramatically, supporting the datacenter ecosystem. This is what happened with coal usage in the 1800s. Microsoft’s CEO Satya Nadella even referenced Jevons Paradox on Sunday on X based on the DeepSeek news. 
Despite this potential for a swell in overall demand for chips and power, Luke is more cautious about expectations for AI infrastructure plays. For example, here’s his take on Nvidia: We wouldn’t get too bullish on AI hardware and semiconductor stocks like Nvidia. Some will likely lose their pricing power as the hardware in this industry becomes increasingly commoditized, leading to lower margins and slower profit growth. That said, it’s far too early to leap to the conclusion that you need to bail on these stocks. For now, a “wait and see” approach is more appropriate. We’re still just scratching the surface There are plenty of additional details about DeepSeek and its impact on U.S. AI stocks that we haven’t had space to cover here. That’s why our experts Louis Navellier, Eric Fry, and Luke Lango just sat down with our Editor-in-Chief and fellow Digest writer, Luis Hernandez, for a roundtable discussion. While you’re likely familiar with Louis, Eric, and Luke, you may not know that in December, they created their AI Revolution Portfolio together. This is a “best of the best” collection of AI stocks based on our experts’ respective approaches to picking market winners. The portfolio represents what Louis, Eric, and Luke believe are the best ways to profit from the AI revolution over the next 12-26 months. I’ll note that their AI Revolution Portfolio returned more than 21% last year... and continues to outperform the market this year. So, who better to comment on DeepSeek and its impact on AI stocks than these three experts? In their video roundtable discussion below, Louis, Eric, and Luke dive into the vast implications of DeepSeek’s sudden emergence, and most importantly, what you should do about it in your portfolio. Click here (or press the play button on the image below) to watch them break down the winners and losers in this AI revolution, as well as the moves that savvy investors need to make today. 
And to learn more about the “best of the best” AI stocks in their AI Revolution Portfolio, click here. Wrapping up… Wall Street is still getting a handle on what DeepSeek will mean for AI, and the takeaways are likely to shift over the coming weeks/months. But as it looks today, this is bullish for AI stocks and your portfolio. Here’s Luke to take us out: In our view, DeepSeek’s breakthrough will meaningfully accelerate AI model advancement and meaningfully boost the odds that Big Tech companies achieve AGI much sooner than previously thought. We view DeepSeek’s efficiency breakthrough as great news for the industry – and great news for AI stocks, too… So… as it relates to all the fears about DeepSeek’s breakthrough meaning the end of Big Tech or the death of AI… we think those fears couldn’t be further from the truth. Have a good evening, Jeff Remsburg | 
Comments
Post a Comment