rolvram
3 시간 전
Will Chinese DeepSeek Disrupt NVIDIA?
The stock price of NVDA stock is probably the best guide to this answer.
Itai Levitan
Itai Levitan
Saturday, 25/01/2025 | 17:35 GMT-5
2
Will Chinese DeepSeek Disrupt NVIDIA and the American AI Industry?
The global AI race has taken an intriguing turn with the emergence of China's DeepSeek, touted by some as a potential disruptor to NVIDIA's dominance in the AI hardware space. As speculation swirls, stock investors, analysts, and thought leaders are left grappling with the critical question: Is DeepSeek a genuine threat to NVIDIA and the broader American AI industry, or is this a case of overhyped news amplified by interested parties?
The news that DeepSeek's model surpassed OpenAI's o1 in specific reasoning tests has sparked intense debate within the AI community. This development has significant implications for the AI market, potentially impacting the stock prices of companies like NVIDIA and other AI-related companies.
I’ll preface this by saying that I am not a renowned expert in deep technology or AI hardware. I don’t claim to have a definitive answer. Instead, I propose a method to find the answer—not by theorizing endlessly but by observing how the stock market reacts. The stock market, with its collective wisdom and ability to process information, often acts as the closest thing we have to a truth detector.
Let’s explore the context, evaluate the key factors at play, and develop a logical framework to assess the significance of DeepSeek’s emergence.
The Landscape: NVIDIA’s Moat vs. DeepSeek’s Promise
NVIDIA has long been the undisputed leader in AI hardware, leveraging its GPUs and CUDA software ecosystem to dominate the market. Its ecosystem is sticky—developers, hyperscalers, and enterprise clients are deeply entrenched in its platform, creating significant switching costs. This is not unlike Apple’s ecosystem, where loyal users and interconnected services create barriers to exit.
DeepSeek, on the other hand, reportedly offers a more cost-efficient AI solution. The analogy some have drawn is that DeepSeek could be the “cheaper smartphone” to NVIDIA’s “iPhone.” In the smartphone industry, the availability of cheaper alternatives has undeniably shifted market share, particularly in emerging markets. But did this disrupt Apple’s stock price or its dominance among its loyal user base? Not significantly. The same could apply to NVIDIA, depending on whether DeepSeek can replicate NVIDIA’s ecosystem or merely compete on price and performance.
The Role of Market Reactions: Finding the Answer in the Price
Here’s where the stock market comes in. Instead of relying solely on speculative opinions, I suggest that the best way to gauge the impact of DeepSeek is to observe NVIDIA’s stock price. The stock market is often a better truth-teller than any single analysis. It reflects collective sentiment, informed by analysts, insiders, and countless external factors.
What to Watch:
The Reaction to the News
Since the DeepSeek news broke, NVIDIA’s stock has not shown any drastic reaction. It actually rose apx 2.5% after the news came out, if we look at NVDA stock price from the next day it traded, to the close of the week. This lack of significant movement, not to mention lack of stock price decline (crash...?) -- suggests that the market does not yet see DeepSeek as a significant disruptor. If it were, we would expect a sharp and sustained price decline. Who remembers other cases, such as Facebook’s Cambridge Analytica scandal a few years ago and the immediate FB stock decline?
Earnings as a Catalyst
NVIDIA’s upcoming earnings report will be pivotal. Pay close attention to management’s commentary, particularly regarding competitive risks in the AI space. If DeepSeek is a genuine threat, it could show up in NVIDIA’s guidance or performance, especially in its data center segment. I propose observing NVIDIA’s stock price from the time the news broke until two weeks after earnings. This period should provide enough data to assess whether DeepSeek is having a tangible impact, even if will not be talked about. The price knows, and it almost always knows better than you, or anyone else.
Interpreting the Price Action Post Upcoming Nvidia Earnings (Scheduled for 26 Feb 2025): What It Might Tell Us
NVDA stock price is your best guide to the truth
NVDA stock price is your best guide to the "truth"
Scenario 1 for the Upcoming Nvidia Earnings Report: Stock Rises or Remains Stable
If NVIDIA’s stock rises or stays within its current range, the market is likely signaling that DeepSeek is not a major disruptor. Perhaps NVIDIA’s ecosystem is too entrenched, or DeepSeek’s solution doesn’t offer enough differentiation. It’s also possible that other factors, such as AI demand growth or NVIDIA’s new Blackwell chips, outweigh any competitive threats.
Scenario 2 for the Upcoming Nvidia Earnings Report: Stock Declines
If NVIDIA’s stock declines, the reasons warrant closer examination. Is the drop due to bottlenecks in Blackwell chip deliveries, macroeconomic pressures, or broader market trends? Or does it reflect legitimate concerns about DeepSeek? Analysts will need to scrutinize earnings commentary, data center revenue trends, and any shifts in market share. It’s critical to avoid jumping to conclusions without understanding the underlying reasons for the decline.
The Unknowns: Factors and Their Weights
One of the challenges in analyzing stock movements is that we can never know all the factors affecting a stock price, nor the relative weights of those factors. We might think we know, we may hear great explanations, but we don't exacly know. Earnings reports often illustrate this: a company might beat revenue and EPS estimates, yet the stock declines. Analysts scramble to explain why—was it slower sales growth, weaker guidance, or some other reason? The reality is that the stock market processes countless variables simultaneously, and not all of them are visible to us. The granular variables and their weights will never be known to us. But price knows.
The same principle applies here. Even if NVIDIA’s stock remains stable, it’s possible that DeepSeek is a growing threat but not yet significant enough to outweigh other factors. Conversely, if the stock declines, DeepSeek may not be the primary cause. This complexity is why I emphasize looking at the stock price holistically and over time, rather than rushing to judgment based on short-term moves.
Seeking the Deep Behind DeepSeek: Trust the Price, Not the News
DeepSeek’s potential to disrupt NVIDIA and the American AI industry is an open question, but the best way to find an answer is to let the stock market speak. Watch NVIDIA’s price action from the time the news broke until two weeks after its upcoming earnings. If the stock remains stable or rises, it’s likely that DeepSeek is not a significant threat. If it declines, dig deeper into the reasons—whether they’re related to DeepSeek or other factors.
While price action can act as a “truth detector,” it’s not infallible. The market may not immediately price in all information, and short-term movements can be influenced by noise, speculation, or interested parties (e.g., short sellers amplifying fears about DeepSeek). As always, exercise caution and remain open to multiple interpretations. But do follow the price more than anyone's opinion, especially yours.
Ultimately, the price knows. Follow the money. Follow the price and you shall be closer to finding the truth.
So far, here is the most simple price chart of NVDA stock within the past 100 days. I don't see any panic yet from the "truth machine", do you? You can decide for yourself is you see any DeepSeek news related panic. But let's wait till 2 weeks after the next earnings to unveal further "truth".
Whalatane
17 시간 전
U aren't concerned about Deep Seek ?. Their AI advances effectively crater demand for NVDA's chips ( and I own NVDA in my funds ) .
Monday is shaping up unfortunately as a blood bath for AI related US Co's .
Hope I'm wrong ...US Software Co's are apparently the winners here ...so hopefully thats cancels out some of the loss's the AI hardware stocks might experience
DeepSeek, a cutting-edge large language model (LLM), offers significant advancements that help software companies in several ways:
Efficient Code Generation: DeepSeek automates coding tasks, potentially cutting development time by up to 40%1. It supports over 80 programming languages, allowing for versatile code generation across different projects2.
Enhanced Code Quality: The model excels at code review, identifying errors and suggesting real-time optimizations to improve code quality and maintainability1.
Accelerated Debugging: DeepSeek analyzes error logs, detects patterns, and automates fixes, significantly speeding up issue resolution in software development1.
Cost-Effective Performance: With its Mixture-of-Experts (MoE) system, DeepSeek activates only 37 billion of its 671 billion parameters for any task, reducing computational costs while maintaining high performance1.
Improved Problem-Solving: DeepSeek scores 84.1% on GSM8K (problem-solving benchmark), enhancing algorithm design and complex problem-solving capabilities for developers1.
Long Context Processing: The model can handle up to 128K tokens, allowing it to maintain coherence across large codebases and complex systems1.
Open-Source Accessibility: As an open-source model, DeepSeek enables smaller teams and organizations to deploy and customize AI tools without significant infrastructure costs1.
Rapid Integration: DeepSeek can be integrated into existing workflows, such as through platforms like Latenode, allowing companies to quickly automate coding tasks within their current processes2.
By leveraging these advancements, software companies can significantly improve their development efficiency, code quality, and problem-solving capabilities while managing costs effectively.
Kiwi
doc2016
2 일 전
how cuda-q could be used in mrna work.......it's essential or the project of such could never be considered.
"asking gemini ai some questions, but thinking add inovio ip of plasmid and synthetic dna with potential of the synthetic dna using msft ip...see below.
"Microsoft's research into DNA data storage involves a fascinating process of converting digital information into the language of life. Here's a simplified breakdown of how they transcribe data into DNA:
1. Encoding Digital Data:
Binary to Base-4: Digital data is typically stored in binary code (0s and 1s). This binary code is converted into a base-4 format, which aligns with the four nucleotide bases in DNA: adenine (A), cytosine (C), guanine (G), and thymine (T).
Example: 00 could be encoded as A, 01 as C, 10 as G, and 11 as T.
2. DNA Synthesis:
Creating DNA Strands: A DNA synthesizer machine uses this base-4 code to chemically create synthetic DNA strands. Each strand represents a portion of the digital data.
Redundancy and Error Correction: Multiple copies of the DNA strands are created, and error-correcting codes are incorporated to ensure data integrity.
3. DNA Storage:
Dried and Stored: The synthesized DNA is dried and stored in small vials, which can be kept at room temperature for long periods.
4. Data Retrieval:
DNA Sequencing: When the data needs to be retrieved, the DNA is rehydrated and sequenced using a DNA sequencer. This machine reads the sequence of nucleotide bases in the DNA strands.
Decoding: The sequenced DNA code is then decoded back into the original binary format, restoring the digital data.
Key Technologies:
DNA Synthesis: Advances in DNA synthesis technology have made it possible to create custom DNA strands with high accuracy and efficiency.
DNA Sequencing: High-throughput DNA sequencing technologies allow for rapid and accurate reading of DNA sequences.
Error Correction: Sophisticated error-correcting codes are used to ensure data integrity during storage and retrieval.
Advantages of DNA Data Storage:
High Density: DNA can store vast amounts of information in a tiny volume.
Durability: DNA is a very stable molecule, capable of preserving information for thousands of years.
Low Energy Consumption: DNA storage requires minimal energy compared to traditional data centers.
Challenges:
Cost: DNA synthesis and sequencing are currently expensive processes.
Speed: Writing and reading data from DNA is slower than traditional storage methods.
Automation: The process of encoding, synthesizing, storing, and retrieving data from DNA needs to be further automated for practical applications.
Microsoft's Progress:
Microsoft has made significant strides in automating the process of DNA data storage and retrieval. They have demonstrated the ability to store and retrieve data in DNA with high accuracy, and they are working to reduce the cost and increase the speed of this technology.
While DNA data storage is still in its early stages, it holds immense potential for long-term archival storage and could revolutionize the way we store and preserve information in the future.
Sources and related content
What Is Digital Data? - Egnyte
www.egnyte.com
A Practical Approach to Ultra Long-Term Data Storage on DNA - Microsoft Research
www.microsoft.com
DNA for Data Storage and Retrieval - Federation of American Scientists
fas.org
DNA Sequencing Fact Sheet - National Human Genome Research Institute
www.genome.gov
Microsoft and University of Washington researchers set record for DNA storage
blogs.microsoft.com
can the microsoft dna transcription method be used to make a vaccine?
While Microsoft's DNA data storage technology and the creation of vaccines both involve DNA, they serve very different purposes and utilize distinct processes.
Here's why Microsoft's DNA transcription method is not directly applicable to vaccine development:
Microsoft's DNA Data Storage:
Focus: Storing digital information (0s and 1s) in the sequence of DNA bases (A, C, G, T).
Process: Encoding digital data into a base-4 format, synthesizing DNA strands that represent the data, and using sequencing to retrieve the information.
Goal: Long-term archival storage of vast amounts of data.
Vaccine Development:
Focus: Triggering an immune response in the body to protect against diseases.
Traditional Methods: Often involve using weakened or inactive viruses or viral proteins to stimulate the immune system.
mRNA Vaccines: Newer mRNA vaccines deliver genetic instructions (mRNA) to cells, prompting them to produce a viral protein that triggers an immune response.
Goal: Disease prevention and immunity.
Key Differences:
Information vs. Biological Function: DNA data storage focuses on encoding and storing information, while vaccines utilize DNA or RNA to trigger biological processes in the body.
Synthetic vs. Natural: Microsoft's method uses synthetic DNA created in a lab, while vaccines often utilize natural or modified biological components.
Data Retrieval vs. Immune Response: DNA data storage aims to retrieve the stored information, while vaccines aim to stimulate a lasting immune response.
Potential Indirect Connections:
While not directly applicable, Microsoft's research in DNA manipulation could indirectly contribute to vaccine development in the future:
DNA-based Vaccines: Some vaccines under development use DNA to deliver genetic instructions to cells, similar to mRNA vaccines. Microsoft's expertise in DNA synthesis and manipulation could potentially be helpful in this area.
Data Storage for Vaccine Development: DNA data storage could be used to store vast amounts of genomic data and research related to vaccine development, aiding in the identification of new targets and the design of more effective vaccines.
In Summary:
Microsoft's DNA transcription method for data storage is not directly used to make vaccines. However, their research in DNA technologies could indirectly contribute to advancements in vaccine development in the future.""
rolvram
2 일 전
Meta to spend up to $65 bln this year to power AI goals, Zuckerberg says
10:01:52 AM ET, 01/24/2025 - Reuters
(Adds details and background throughout)
Jan 24 (Reuters) - Meta Platforms plans to spend between $60 billion and $65 billion this year to build out AI infrastructure, CEO Mark Zuckerberg said on Friday, joining a wave of Big Tech firms unveiling hefty investments to capitalize on the technology.
As part of the investment, Meta will build a more than 2-gigawatt data center that would be large enough to cover a significant part of Manhattan. The company — one of the largest customers of Nvidia's coveted artificial intelligence chips — plans to end the year with more than 1.3 million graphics processors.
"This will be a defining year for AI," Zuckerberg said in a Facebook post. "This is a massive effort, and over the coming years it will drive our core products and business."
Zuckerberg expects Meta's AI assistant — available across its services, including Facebook and Instagram — to serve more than 1 billion people in 2025, while its open-source Llama 4 would become the "leading state-of-the-art model".
Shares of the company were 1.6% higher in early trading.
Big technology companies have been investing tens of billions of dollars to develop AI-related infrastructure after the meteoric success of OpenAI's ChatGPT highlighted the potential for the technology.
U.S. President Donald Trump on Tuesday announced that OpenAI, SoftBank Group and Oracle will form a venture called Stargate and invest $500 billion in AI infrastructure across the United States.
Earlier this month, Microsoft said it was planning to invest about $80 billion in fiscal 2025 to develop data centers, while Amazon.com has said its capital spending for 2025 would be higher than an estimated $75 billion in 2024.
Meta's planned capital spending of up to $65 billion would mark a significant jump from its estimated capital spending of $38 billion to $40 billion for last year.
As part of the AI efforts, the company said it would build an AI engineer that will start contributing increasing amounts of code to its research and design efforts. It will also continue to grow the teams working on AI services.
doc2016
2 일 전
lockheed martin had their own stargate which some considered spooky occurences in the sky.
gemini ai says, "It seems you might be thinking of two different "Stargate" projects. Let's clear up the confusion:
1. The Historical Stargate Project:
U.S. Army Initiative: This was a real-life government-funded program that ran from the 1970s to 1995, exploring psychic phenomena like remote viewing for potential military and intelligence applications.
Declassified: The project was declassified in 1995, and its findings were largely inconclusive.
Lockheed Martin's Involvement: There is no public evidence to suggest that Lockheed Martin was directly involved in this historical Stargate Project.
2. The Recent Stargate Project (AI Infrastructure):
AI Data Centers: This is a new initiative announced in January 2025 by OpenAI, Oracle, Softbank, and MGX to build a massive network of AI-focused data centers in the United States.
Lockheed Martin's Role: As of now, there is no indication of Lockheed Martin's direct involvement in this AI-focused Stargate project.
Possible Indirect Connections:
While Lockheed Martin might not be directly involved in either Stargate project, there are potential indirect connections:
AI and Defense: Lockheed Martin is a major defense contractor heavily invested in AI for applications like autonomous systems, cybersecurity, and intelligence analysis. They could potentially benefit from or contribute to the AI infrastructure created by the new Stargate project.
Advanced Technologies: Lockheed Martin has a history of working on advanced technologies, some of which might have overlapped with the interests of the historical Stargate Project (e.g., sensor technology, signal processing). However, any connection remains speculative.
In conclusion:
It's important to distinguish between the two Stargate projects. While Lockheed Martin might have tangential connections to the domains of both projects, there is no public information confirming their direct involvement in either initiative."
doc2016
2 일 전
to supplement the discussion:
1. softbank and son bring the potential involvement of other vision fund 1/2 investors, including oil rich at this point in time emirates, but who are likely looking for life after oil
2. lockheed martin's quest for fusion reactor could be greatly aided by supercoputer sims of magnetic field containment around the plasma, pusle rate of the laser, quantum computer aided materials research, looking for new qubit materials, optimization of networking, routing, authentication for same, etc.
3. ambani owns reliance which bought radisys which bought texas micro which worked on 5-9's monitoring and failover work with sunmicro, msft, hp ...5-9's is important and even more so with the kind of network speed required for the low latency work of hybrid data center.
4. even knowing the plasma field might require use of microwave/infrared lasers and sensors to optimally manage the whole containment. no leakage can be allowed they say.
Jetmek_03052
2 일 전
Here's the ACTUAL words said in this "Stargate" AI investment article:
The venture involves OpenAI, SoftBank, and Oracle, and starts with an investment of $100 billion.
On the second day of his second presidency, Donald Trump announced the creation of a joint venture with OpenAI, SoftBank, and Oracle to build large-scale AI data centers in the United States. The new company, named Stargate, will be funded with a $100 billion initial investment, which could grow to $500 billion within the next four years.
This is the beginning of the investment. Who KNOWS where the rest of the money will come from??? There's more than a few company's that might jump into it. Four years leaves a lot of time for investment.
Granted, it may not get all the way to half a trillion dollars. And this "Stargate" is not a new idea. Also, it is well known that Musk is at odds with OpenAI. I'll wager that Musk was approached about investment and once he heard that OpenAI was a part of it, he refused. This is one of the reasons he seems doubtful about the final investment number.
As usual, you jump on something and force negativity upon it.
Hot Features
프리미엄
실시간 차트, 분석 도구 및 가격을 받으려면 등록하세요.
DiscoverGold
16 시간 전
NASDAQ2020
1 일 전
JJ8
2 일 전
IanFromSI
2 일 전
BottomBounce
3 일 전
