Nvidia and Meta Analysis Today 29/1: DeepSeek shakes up AI

The announcement of the new Chinese AI app, DeepSeek R1, shook the tech market and caused a 17.0% drop in the shares of the current leader in the AI hardware industry, Nvidia (NASDAQ: NVDA).

This shocking news has 2 important implications:

• The Chinese start-up that developed the model was able to complete the project despite U.S. restrictions on the sale of high-tech microprocessors to the Asian giant.
• According to the given information, the costs incurred in the development and training of the model represented only a small fraction of what U.S. tech companies are investing, thanks to a more modular approach that is less intensive in hardware and energy use.

Additionally, the model is open-source and so far is free …

However, not all AI stocks suffered equally. Meta Platforms (NASDAQ: META) has reached new all-time highs and remains strong in its social media advertising business.

If what DeepSeek announced is true. Are we spending too much on AI and are the valuations of the entire industry overestimated? Are we getting close to the end of a bubble or did the market overreact?

DeepSeek: The announcement that shook the AI industry

The introduction of the DeepSeek R1 model by a Chinese start-up is causing significant repercussions in the global tech landscape, particularly among American tech giants like Nvidia, Meta, Google, and OpenAI.

This sophisticated AI system has demonstrated unprecedented capabilities in data analysis, enabling faster and more accurate decision-making processes compared to existing models like ChatGPT.

Why are investors selling their AI stocks?

The recent introduction of the DeepSeek R1 model has caused a significant drop in AI stocks in the United States, both in hardware and in the provision of infrastructure for data centers and energy.

Investors are concerned about the potential disruption that this advanced AI technology could cause in the market. A free and open-source model could undermine the profitability of American tech companies operating in this industry. They would no longer be exceptional as they have been until now!

This situation has prompted a reassess of technology valuations for companies that have heavily invested in AI development. What if it were possible to develop a model without spending as much money as China shows?

As market dynamics evolve, American companies are now under pressure to innovate quickly, which could lead to significant adjustments in AI stock valuations.

Is the spending model on AI correct?

“Necessity is the mother of invention,” says a well-known proverb… It seems that this applies perfectly to what happened with the announcement of DeepSeek R1.

A critical aspect driven by the emergence of this new AI model is the evaluation of AI expenses. While significant investments have been channelled into the research and development of this technology, the creators of DeepSeek R1 claim to have succeeded at just a tiny fraction of the U.S. costs.

Is it true? Will the world have access to all the project’s information?

Surely not. This will become a state secret in China. But, what cannot be denied under any circumstances is that the available information raises questions about whether American tech companies can maintain a competitive edge without a radical rethinking of their AI strategies and resource allocation.

Nvidia: The leader falls hard against the Chinese outsider

To appreciate the impact of the announcement about DeepSeek R1, let’s take a look at Nvidia’s daily chart:

• On the session of January 27, NVDA opened with a bearish gap of no less than 12.5%. The decline of the stock extended throughout the trading day, ultimately closing with an accumulated drop of 17.0%. This is the largest one-day stock market crash for a single stock, surpassing a loss of more than 500 billion dollars in market capitalization.
• For the session on January 28, the stock was trading at $121.07 after an hour of operation, so it seems to have found support in this area. Now, NVDA is trading 11.4% below its 50-day moving average.
• The announcement of DeepSeek R1 is not a cause, it is a catalyst for the belief that, perhaps, the action has gone too far. Since June 2024, the growth of the stock price had been slowing down…
• The stock is limited on the upside by a resistance level of $152.89 (November 21, 2024) and an all-time high of $153.13 (January 13, 2025).
• The next critical level is around $140.76 and has already been surpassed.
• At the support level, the next relevant value is around $97.40 (March 8, 2024).
• The RSI indicator is quite close to the extreme oversold zone with a level of 37.11.

Meta Platforms: The stock reaches new all-time highs

Meta has been doing much better with the recent news. Next, we present their daily chart:

• Although META opened on the 27th with a bearish gap of 3.2%, it was able to recover throughout the session and close with an increase of 5.3% from the opening.
• On the 28th, the stock reached a new all-time high of $675.48, before partially retreating to $661.50 by the end of the first hour of trading.
• The company fluctuated within a sideways range between February and August 2024, but has now resumed the long-term upward trend that began in October 2022. At this moment, it is trading 9.1% above its 50-day moving average.
• The next critical level is above $638.40 (December 11, 2024). Beyond that, we have 2 additional levels at $602.95 (October 7, 2024) and $549.05 (November 21, 2024).
• The RSI indicator is projected into the overbought zone with a value of 68.46.
• At the close of the trading session on January 28, the results for the fourth quarter of 2024 will be published, and all indications are that the sales revenue and earnings per share metrics will continue to exceed analysts’ expectations, just as they have uninterruptedly since the first quarter of 2023. The stock is likely to trend upwards for the rest of the session.
• META is benefiting from the good state of the U.S. economy in terms of better expectations for consumers and investors due to Trump’s new economic policies. This translates into higher advertising revenues.
• The stock could very well project towards $700.00 in the short term.

Regarding its own open-source AI model, Meta plans to launch Llama 4 this year, including advancements in reasoning, video generation, and speech models. The company is also developing enterprise AI agents to deepen interactions and assistance in digital commerce.

Conclusion

The announcement of the new Chinese AI model, DeepSeek R1, is of great importance for the industry and for technological advancement itself. Greater competition always leads, in the long run, to lower costs for the services provided to the end consumer, and that is positive.

These initiatives will help increase the productivity of individuals, companies, and nations that are willing to take advantage of them.

Whether the project’s data and figures are accurate or not will be a matter of great speculation. The announcement is made after the President of the United States himself made public an alliance (Project Stargate) with OpenAI, Oracle, and Softbank, among others, which aims to invest up to 500 billion dollars in the coming years to develop critical infrastructure for the sector.

Moreover, there is an ongoing negotiation between the United States and China regarding the possible ban on TikTok and the threat of tariffs.

So, this news was not announced by chance… Given the nature of the Chinese political system, it is necessary to be cautious with information coming from this country, as everything is strictly controlled by the State.

Regarding Nvidia, this news is likely to dampen investors’ risk appetite in AI and lead the company’s stock towards some level of correction in the short term. They have already fallen by 17%!

If you invest for the long term, this could represent an opportunity to position yourself in this formidable company. That it is possible to develop and train an AI model without making gigantic investments does not mean that the world does not require the chips that Nvidia produces. However, it does mean that the company must diversify its markets and the scope of its products.

Meta Platforms, for its part, has reached a new all-time high and continues to have social media advertising at the heart of its business model. It should not be overlooked that Meta had already announced that it would be moving towards an open-source AI model with Llama… So that it can continue creating ecosystems and tools that attract the public as a lever to keep selling ads and other digital products.

More than a reason for panic, the Chinese announcement should be taken with great respect, immense curiosity, and, above all, as a wake-up call about the future. The United States is not alone in the technology arena, and the future can always bring us unexpected surprises!

Frequently Asked Questions

What is the DeepSeek R1 model?

The DeepSeek R1 model is an advanced artificial intelligence system developed by a Chinese tech start-up, free and open-source, which would have the ability to analyse large amounts of data more efficiently than existing models, such as OpenAI’s ChatGPT, Meta’s Llama, or Google’s Gemini.

Why are some AI stocks falling?

AI stocks are facing declines due to investors’ concerns about the impact of this technological milestone on competition, as well as doubts about the efficiency of the current AI investment model and the sector’s stock valuations.

What can be expected from American tech companies?

After evaluating the true scope and implications of the DeepSeek R1 model, U.S. tech companies will rethink their investment strategies to adapt to this competitor and focus on more efficient innovation.

Ready to trade our Forex daily forecast? We’ve shortlisted the best forex broker list for you to check out.