Nvidia Fuels AI Race: $2B Bets Spark Tech Stock Surge

Nvidia’s Strategic AI Investments Ignite Market Buzz

While global headlines focused on geopolitical tensions, significant developments in the artificial intelligence sector have been reshaping the investment landscape. Nvidia, a leader in AI technology, has made substantial investments, totaling $2 billion, in two key companies: Nebius and Marll Technologies. These moves signal a strong belief in the future growth of AI infrastructure and specialized chip development.

Nebius Secures Major AI Deals, Eyes Massive Revenue Growth

Nebius, a company focused on providing AI infrastructure for secure and regulated industries like finance and healthcare, has received a major endorsement. Nvidia is investing $2 billion in Nebius as part of a partnership aimed at deploying over 5 gigawatts of Nvidia systems by 2030. This amount of power is enough to supply nearly 4 million homes. Nebius will also gain early access to Nvidia’s latest Vera Rubin platform. Following this, Nebius announced a significant 5-year deal with Meta Platforms, potentially worth up to $27 billion. This agreement involves supplying Meta with 12 billion compute capacity by 2027, including early deployments of the Vera Rubin GPUs. Meta also holds an option to purchase an additional $15 billion. Analysts predict these deals could boost Nebius’s annual recurring revenue by 300-400%, reaching an estimated $4 to $6 billion once fully operational. Nebius management projects total annual recurring revenue of $7 to $9 billion by the end of 2026, representing more than a fivefold increase year-over-year.

Nebius Stock Valuation and Investor Considerations

Based on these developments, financial models suggest Nebius’s stock could be significantly undervalued, with a fair value estimated at over $600 per share, implying an 80% upside potential. However, investors should note that Nebius remains a small and volatile stock with a high price-to-earnings ratio, making it a higher-risk investment.

Microsoft’s Long-Term Trend and Buying Opportunity

In a notable technical development, Microsoft’s stock has touched its 200-day moving average for the first time in over 13 years. This average represents a stock’s average closing price over the past 4 years, acting as a long-term trend indicator. For strong companies, this line often serves as a support level where buyers tend to step in. A decline to this level for a quality stock like Microsoft can signal a rare long-term buying opportunity, rather than a temporary dip. This strategy aligns with the investment philosophy of the late Charlie Munger, who believed that consistently buying high-quality stocks at their 200-day moving average could lead to outperforming the S&P 500 over time.

Microsoft’s Performance Amidst Market Volatility

Microsoft’s stock has seen a decline of over 20% year-to-date, making it one of the weaker performers among the ‘Magnificent 7’ tech giants, second only to Tesla. However, unlike Tesla, which has experienced declining revenues and profits, Microsoft has grown its revenues, gross profits, and earnings per share by approximately 15% in the past year, from a much larger base. Financial models estimate Microsoft’s fair value at around $450 per share, with some analysts setting targets above $500. This suggests the stock is currently undervalued by 17% to 25%, offering potential upside of 20% to 33%.

Marll Technologies Partners with Nvidia for AI Advancement

Nvidia’s strategic investments extend to Marll Technologies, with a $2 billion investment and a partnership to integrate Marll’s custom chips into Nvidia’s ecosystem via NVLink Fusion. This technology allows chips from other companies to connect seamlessly with Nvidia’s GPUs at full speed. This enables Nvidia to benefit from infrastructure components like CPUs and networking gear, even when data centers use chips from competitors like Intel or AMD. Marll will develop custom AI chips, known as XPUs, and high-speed networking equipment that will connect to Nvidia’s systems. This collaboration also includes advancements in silicon photonics and optical interconnects, a technology that uses light to transfer data, potentially speeding up AI operations.

Marll’s Strategic Position Post-Acquisition

This partnership follows Marll’s recent $3.25 billion acquisition of Celestial AI, a company specializing in photonic fabric. By integrating cutting-edge optical technology, Marll is positioning itself as a key player within Nvidia’s AI infrastructure strategy.

ARM Enters the AI Chip Race with New CPU Design

ARM, traditionally a designer of licensed chip architectures, has announced its first self-branded chip, the AGI CPU. Designed for AI inference in data centers, this chip can support up to 136 cores per CPU. Meta Platforms is the initial customer, with OpenAI, Cerrus, and Cloudflare also expected to use the new chip. However, ARM’s move into direct chip competition could create tension with its major customers, including Qualcomm, Amazon, and Microsoft, who license ARM’s architecture. The long-term impact will depend on the chip’s performance and its ability to compete effectively once it reaches mass production.

Anthropic Secures Major Compute Deals, Accelerates AI Development

Anthropic, known for its Claude family of AI models, has secured significant compute resources through new deals with Google and Broadcom. The company will gain access to approximately 3.5 gigawatts of next-generation TPU (Tensor Processing Unit) capacity starting in 2027, in addition to 1 gigawatt of Google’s TPUs coming online this year. This combined 4.5 gigawatts is enough to power over 3 million homes. Broadcom will supply the custom TPUs and networking equipment under a long-term agreement with Google through 2030, ensuring a stable revenue stream.

Anthropic’s Rapid Revenue Growth and Investment Opportunities

Anthropic’s revenue run rate has surged to over $30 billion annually, a dramatic increase from around $9 billion at the end of 2025. This rapid growth positions Anthropic as a major player in the AI field. For retail investors, exposure to Anthropic is available through the Fundrise Innovation Fund (VCX), which recently began trading.

Project Glasswing: Securing the Future with Advanced AI

Anthropic is developing a new, highly powerful AI model called Claude Mythos. Due to its potential risks, Anthropic has formed Project Glasswing with industry leaders like Crowdstrike, Palo Alto Networks, Broadcom, and Nvidia. This coalition will use Mythos defensively to identify and patch critical vulnerabilities in their own systems before they can be exploited by malicious actors. This initiative highlights Anthropic’s evolving role as a critical infrastructure and security partner, while providing cybersecurity firms with a significant advantage in combating AI-powered threats.

Google DeepMind’s Turboquant Breakthrough Impacts Memory Stocks

Google DeepMind has unveiled Turboquant, a new compression technique that significantly reduces the memory required for AI models’ KV cache. This innovation can shrink memory needs by up to sixfold with no loss in accuracy and potentially speed up AI calculations by eight times. The announcement led to a sharp drop in memory stocks, with Samsung, SK Hynix, and Micron falling 5-7%, and SanDisk experiencing a double-digit decline. This reaction stems from concerns that AI data centers might require fewer memory chips than previously anticipated.

Jevons Paradox and the Memory Market Outlook

However, the situation may mirror Jevons paradox, where increased efficiency leads to greater overall demand. Similar to how Nvidia’s stock initially dropped due to efficiency gains but ultimately saw demand soar, Turboquant could drive increased demand for memory as AI models become more accessible and widely adopted. Micron reported exceptional earnings, with revenues up 75% quarter-over-quarter and 196% year-over-year, and earnings exploding by over 700%. The company guided for further 40% growth in the next quarter. The launch of the Roundhill Memory ETF (DRAM) further underscores the growing importance of memory as a critical, yet constrained, component in the AI supply chain.

SpaceX Eyes Historic IPO Amidst Ambitious Expansion Plans

SpaceX is preparing for what could be the largest IPO in market history, with potential valuations ranging from $1.5 to $2 trillion, though Elon Musk has downplayed these figures. The company’s public offering would encompass its diverse operations, including reusable rockets, Starlink satellite internet, Grok AI models, and its social media platform, X. A key long-term objective is the development of orbital data centers, which would leverage space-based computing to overcome Earth’s limitations.

Terrafab Project and Intel Partnership

To support these ambitions, SpaceX is planning the Terrafab manufacturing complex in Austin, Texas, a $20-$25 billion project aimed at producing custom AI chips for Tesla, SpaceX, and future orbital data centers. Intel has signed on as the manufacturing partner, providing chip design, fabrication, and packaging expertise. The Terrafab project aims to produce up to 1 terawatt of compute capacity annually, a scale significantly larger than the current global AI chip production.

Market Impact and Investor Considerations

Geopolitical Volatility and Market Sentiment

The recent geopolitical events, particularly concerning Iran, initially heightened market fear, as indicated by CNN’s Fear and Greed Index remaining in ‘extreme fear’ for an extended period. However, a swift de-escalation led to a market rally, with oil prices dropping 15%. This highlights the market’s sensitivity to geopolitical news and the potential for rapid shifts in investor sentiment. Tracking indicators like the Fear and Greed Index can help investors identify opportunities when others are overly fearful.

AI Infrastructure and Chip Dominance

Nvidia’s significant investments in Nebius and Marll Technologies underscore the critical role of AI infrastructure and specialized chip manufacturing. Companies like Nebius, with its focus on sovereign AI, and Marll, with its custom chip designs and integration capabilities, are poised for substantial growth. Investors seeking exposure to this trend might consider companies involved in AI hardware, cloud computing, and semiconductor technology.

Long-Term Value in Quality Tech Stocks

Microsoft’s stock approaching its 200-day moving average presents a potential long-term buying opportunity for investors who believe in the company’s continued growth. Despite short-term market fluctuations, its strong financial performance and strategic positioning in AI suggest resilience. The significant undervaluation indicated by financial models further supports this view.

The Evolving Memory Market

The breakthrough in memory compression technology, Turboquant, initially caused a sell-off in memory stocks. However, the potential for increased overall demand due to greater efficiency aligns with Jevons paradox. The strong performance of companies like Micron and the introduction of a dedicated memory ETF (DRAM) suggest that memory remains a crucial and potentially profitable segment of the AI supply chain.

SpaceX’s Ambitious Future and IPO Potential

SpaceX’s potential IPO and its ambitious plans, including orbital data centers and large-scale chip manufacturing through Terrafab, represent a significant long-term growth narrative. While valuations are speculative, the company’s integrated approach across rockets, satellite internet, AI, and social media positions it for substantial future impact. Investors should monitor the progress of these ventures and the eventual IPO details.

Cybersecurity’s AI-Driven Evolution

Anthropic’s advancements in AI models and its role in Project Glasswing highlight the increasing importance of AI in cybersecurity. Companies like Crowdstrike and Palo Alto Networks, by gaining early access to powerful AI tools for threat detection, are strengthening their competitive positions against emerging AI-powered cyber threats.

Where to Find More Information

For investors looking to analyze companies based on valuation metrics like discounted cash flow models and analyst estimates, tools like Simply Wall Street can be invaluable. The platform offers a stock screener and watch list features to help identify potentially undervalued companies.

Source: How I'm Investing After The Stock Market Drop (YouTube)