US DeepSeek takes root. NVDA, OpenAI Doomed?
Blazing DeepSeek
The emergence of DeepSeek, a Chinese AI developer has caused deep distress to both $NVIDIA(NVDA)$ and OpenAI.
It resulted in them losing billions in value, -$600 billion to be exact in Nvidia’s case.
Based on past month trading sessions, Nvidia has lost -16.35% to its stock price. (see below)
Although it has recovered (+5.35%) on Wed, 05 Feb 2025 - it is forecasted to be volatile in the near term because (a) it has not reported its quarterly earnings to quell market uneasiness with it and (b) Deepseek is here to stay.
Certainly, a lot of US big techs have been trying (very hard!) to “brush off” DeepSeek emergence : (see above)
OpenAI CEO Sam Altman, $Meta Platforms, Inc.(META)$ CEO Mark Zuckerberg and $Microsoft(MSFT)$ CEO Satya Nadella have all appeared largely unconcerned about the new AI model in recent days.
Meta Platform.
-
On Wed, 29 Jan 2025, during Meta’s quarterly earnings conference, CEO Zuckerberg voiced confidence in Meta’s plans to continue investing heavily in AI infrastructure in the wake of DeepSeek’s emergence.
-
For 2025, Meta plans to spend between $60 to $65 billions on capital expenditures as it focuses on AI.
-
Zuckerberg’s saving grace was admittance that it is probably too early to really have a strong opinion on what this means for the trajectory around infrastructure and capex and things like that.
-
“There are a bunch of AI-trends that are happening here all at once.
-
Meta Platform is probably the only Mag 7 that has escaped DeepSeek’s Monday crash relatively unscathed.
-
It has since made further inroads in their stock price performances.
Microsoft.
-
Also on Wednesday during Microsoft’s earnings call, CEO Satya Nadella struck a positive tone when discussing DeepSeek.
-
His company announced plans earlier this month to spend $80 billion building out data centers in 2025.
-
He thought DeepSeek has had some real innovations and now that it's widely available, customers benefit most, right?”
-
Comparatively speaking, MSFT did not perform as well as META.
-
It fell, rose and fell again in quick succession.
-
It is still languishing as of Mon, 3 Feb 2025 because according to Wall Street, its quarterly earnings have not been ideal especially its Azure, cloud platform.
-
On 29 Jan 2025 (Wed), Reuters has reported that MSFT has made Chinese startup DeepSeek's R1 AI model available on MSFT’s Azure cloud computing platform and GitHub tool for developers.
-
DeepSeek joins more than 1,800 AI models that Microsoft is offering in its model catalog.
-
Within a week of launch, DeepSeek has overtaken US rival ChatGPT in downloads from Apple's App Store.
-
This comes as Microsoft has been looking to reduce its dependence on ChatGPT maker OpenAI.
-
MSFT has been working to add internal & 3rd-party AI models to power its flagship AI product Microsoft 365 Copilot
-
The main “drawback” to MSFT’s DeepSeek offering is data will be hosted in DeepSeek China’s server.
-
MSFT has announced that customers would soon be able to run the R1 model locally on their Copilot+ PCs, a move that could potentially ease privacy and data-sharing concerns over the use of the model.
OpenAI.
-
On Thu, 30 Jan 2025 Altman acknowledged that DeepSeek “did a couple of really nice things,” but downplayed its overall impact, calling it “wildly overstated”.
-
During a Washington DC event, Altman said DeepSeek is a model at a capability level that OpenAI had quite some time ago and claimed he knows how to run a cheap model.
My thots: Why doesn’t Altman do it ? Instead, he’s been canvasing for billions of dollars of funding while the plan to split OpenAI into a “for profit” and “non-profit” entities, continues to loom.
-
Is it a case of money grab (just like his mentor Tesla CEO, Musk?) ?
DeepSeek: Here To Stay ?
It looks like DeepSeek is not going away anytime soon. (see above)
On Thu, 30 Jan 2025, Cerebras Systems announced that it would host DeepSeek’s breakthrough R1 AI model on US servers, that promises speed of up to 57x faster than GPU-based solutions.
All the while keeping sensitive data within American borders.
The move comes amid growing concerns about (a) China’s rapid AI advancement and (b) data privacy.
Cerebras (a chip startup) will deploy a 70-billion-parameter version of DeepSeek-R1 running on its proprietary wafer-scale hardware, with lighting speed response that is second to none.
It delivers 1,600 tokens per second, a dramatic improvement over traditional GPU implementations that have struggled with newer “reasoning” AI models. (see above)
DeepSeek Reshaping AI landscape.
According to Cerebras, Senior Executive, James Wang:
-
AI reasoning tools will change how people work and they will become part of every knowledge worker's routine.
-
DeepSeek’s appearance has raised questions about Nvidia’s AI supremacy.
-
With Cerebras’s solution offering it addressed 2 key concerns inherent with DeepSeek - (1) computational demands of advanced AI models (see below) and (2) data sovereignty.
-
Cerebras’s implementation of DeepSeek-R1 is executed entirely on US grounds vs using current DeepSeek’s API (a popular download & execute) where data is sent back to China and stored.
-
Cerebras’ hosting solution offers American companies a way to leverage these advances while maintaining data control.
DeepSeek vs OpenAI
-
From 30 Jan 2025, Cerebras’s hosted DeepSeek service will be available through a developer preview.
-
It will initially be free-of-charge, with Cerebras planning to implement API access controls due to strong early demand.
-
Looking at the Cerebras-DeepSeek performance against OpenAI in 3 categories of (a) Question answering, (b) Mathematics and (c) Coding, it is evident which AI solution is more superior.
In a reactionary-attempt to stall China’s AI progress, quite likely that US will further tighten AI-chips export restrictions, hurting Nvidia again in the process
Is this enough a prompt to stall retail investors from buying the dip when it comes to Nvidia ?
Must Read: Click on below titles to access. Repost to share, Like as encouragement ok. Thanks.
-
Do you think DeepSeek will eventually list on US stock exchange ?
-
Do you think Cerebras will take this opportunity to list on Nasdaq ?
If you find this post interesting, give it wings! ️ Repost and share the insights ?
Do consider “Follow me” and get firsthand read of my daily new post. Thank you.
Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.
Pls "Re-post" so that more get to know. Tks! Rating is important (to me).
Consider "Follow me" and get first hand read of my Daily new posts? Thanks!). Tks!!
Great article, would you like to share it?