Oracle's AI-Focused Forecast Boosts Nvidia, TSMC and Other Chipmakers

Oracle projects its cloud infrastructure revenue to surge to $114 billion by fiscal 2030, driving strong gains across AI-related stocks. Nvidia, a key supplier of chips for AI data centers, climbed 4%, while other chipmakers like Advanced Micro Devices and Micron Technology advanced 2% and 4%, respectively. Taiwan Semiconductor Manufacturing Co., which produces chips for Nvidia, rose 4% after reporting a 34% jump in August sales. Server makers Super Micro Computer and Dell Technologies each rose 2%, supported by their role in assembling Nvidia-powered systems.

Oracle has projected its cloud infrastructure revenue to surge to $114 billion by fiscal 2030, driving strong gains across AI-related stocks. This projection comes as AI continues to revolutionize various industries, with a significant demand for advanced computing capabilities.

Key players in the AI semiconductor sector have seen their stocks rise in response to Oracle's forecast. Nvidia, a leading supplier of AI data center chips, climbed 4% following the announcement. This increase underscores Nvidia's pivotal role in the AI ecosystem, as its GPUs are essential for large-scale AI applications. Other chipmakers, such as Advanced Micro Devices and Micron Technology, also saw their stocks rise by 2% and 4%, respectively, indicating broader market confidence in the AI sector.

Taiwan Semiconductor Manufacturing Co. (TSMC), which produces chips for Nvidia and other AI chipmakers, rose 4% after reporting a 34% jump in August sales. This growth can be attributed to the strong demand for AI applications, as highlighted by TSMC's second-highest monthly sales ever in August. The company's 3-nanometer and 5-nanometer processes, which are crucial for advanced AI and high-performance computing, are driving this growth.

Server makers Super Micro Computer and Dell Technologies each rose 2%, supported by their role in assembling Nvidia-powered systems. This increase reflects the growing demand for AI infrastructure, as these companies provide the hardware platforms needed to run AI workloads.

The AI semiconductor race has intensified in 2025, with Broadcom (AVGO) and Nvidia (NVDA) emerging as two of the most formidable contenders. While Nvidia has long dominated the AI chip market, Broadcom’s aggressive R&D investments, strategic product launches, and growing traction with hyperscalers are challenging the status quo. This analysis evaluates whether AVGO is now the superior long-term investment in the AI semiconductor sector, leveraging recent financial performance, innovation pipelines, and competitive dynamics Broadcom's AI Momentum and Strategic Positioning in the Semiconductor Sector^[1].

Broadcom’s AI Momentum: A Surge in Revenue and Custom Solutions
Broadcom’s Q3 2025 results underscored its rapid ascent in the AI semiconductor space. The company reported a record $16 billion in revenue, with AI semiconductor sales surging 63% year-over-year to $5.2 billion, driven by the XPU segment [1]. This growth is fueled by demand for custom AI accelerators from hyperscalers like Google and Meta, which are increasingly prioritizing compute self-sufficiency [1]. Notably, Broadcom secured over $10 billion in orders for AI racks based on its XPUs, prompting a revised and optimistic outlook for fiscal 2026 AI revenue [1].

Broadcom’s competitive edge lies in its vertically integrated approach. The company has launched advanced networking solutions such as Tomahawk 6 (102 terabit per second Ethernet switches) and Jericho 4 (Ethernet fabric routers), which address latency and scalability challenges in large-scale AI deployments [1]. These innovations position Broadcom as a one-stop shop for AI infrastructure, combining custom ASICs with cutting-edge networking hardware. Additionally, Broadcom’s infrastructure software segment, including VMware, grew 17% year-over-year to $6.8 billion in Q3 2025, with the launch of VMware Cloud Foundation 9.0 offering an on-premises alternative to public cloud providers [1].

Nvidia’s Dominance: Ecosystem Leadership and Manufacturing Expansion
Nvidia remains the gold standard in AI semiconductors, with Q2 2026 revenue reaching $46.7 billion—a 56% year-over-year increase—driven by $41.1 billion in data center operations [2]. Its H100 and H200 GPUs continue to power large language models and generative AI applications, despite U.S. export restrictions limiting H20’s availability in China [2]. Nvidia’s gross margins (72.7% non-GAAP) and free cash flow ($39.58 billion in H1 2026) further solidify its financial strength [2].

The company’s strategic initiatives in 2025 highlight its ecosystem-centric approach. Nvidia is building AI supercomputer manufacturing plants in the U.S. with partners like Foxconn and Wistron, aiming to produce half a trillion dollars of AI infrastructure domestically within four years [1]. Innovations like NVLink Fusion—a flexible interconnect enabling third-party CPUs and accelerators to integrate with Nvidia GPUs—underscore its vision of an open AI ecosystem [4]. Meanwhile, the DGX Spark workstation democratizes access to data center-level AI performance, targeting developers and researchers [4].

Comparative Analysis: AVGO vs. NVDA
While both companies are thriving, their strategies diverge. Broadcom’s focus on custom ASICs and networking solutions is resonating with hyperscalers seeking tailored infrastructure, allowing it to chip away at Nvidia’s market share [4]. However, Nvidia’s ecosystem dominance—bolstered by software tools like CUDA, partnerships with cloud providers, and a robust developer community—remains a formidable moat.

Financially, Nvidia’s scale is unmatched. Its AI revenue ($41.1 billion in Q2 2026) dwarfs Broadcom’s ($5.2 billion in Q3 2025), and its 72.7% gross margin [2] far exceeds industry averages. Broadcom, meanwhile, is growing at a faster rate (63% YoY AI revenue vs. Nvidia’s 56% [1][2]), but its $16 billion total revenue pales in comparison to Nvidia’s $46.7 billion.

Broadcom’s VMware integration and Ethernet innovations offer a unique value proposition, particularly for enterprises seeking hybrid cloud solutions. Yet, Nvidia’s recent forays into quantum computing and its global manufacturing expansion [3] suggest a broader long-term vision.

Long-Term Outlook: Which Is the Better Bet?
Broadcom’s momentum is undeniable, but its AI semiconductor revenue still represents a fraction of Nvidia’s. For AVGO to overtake NVDA, it must sustain its growth trajectory while scaling its software ecosystem and expanding into new markets like automotive and edge computing. Conversely, Nvidia’s challenges—geopolitical headwinds in China and rising competition from AMD and Intel—could create opportunities for Broadcom to gain ground.

Investors should also consider valuation metrics. At current price-to-sales ratios, Broadcom appears more affordable than Nvidia, which trades at a premium due to its market leadership. However, Nvidia’s ability to reinvest cash flow into R&D and strategic acquisitions (e.g., its recent quantum computing investments [3]) could justify its higher valuation.

Conclusion
Broadcom’s AI momentum is a compelling story, driven by custom ASICs, networking innovations, and VMware’s growth. However, Nvidia’s ecosystem dominance, financial strength, and strategic foresight make it the more resilient long-term play in the AI semiconductor race. While AVGO is undeniably gaining traction, investors seeking a “sure bet” may still lean toward NVDA—unless Broadcom can close the gap in software integration and global manufacturing.

Broadcom's AI Momentum and Strategic Positioning in the Semiconductor Sector^[1] Broadcom (AVGO) challenges Nvidia (NVDA) in 2025 AI semiconductor race with 63% YoY AI chip revenue growth ($5.2B) and custom XPU traction.
TSMC reports second-highest sales in August^[2] Nvidia maintains dominance via $41.1B data center revenue, 72.7% margins, and ecosystem leadership (CUDA, NVLink Fusion, DGX Spark).
Nvidia's global expansion: AI factories, NVLink Fusion, AI supercomputers and more.
Nvidia's Global Expansion: AI Factories, NVLink Fusion, AI Supercomputers and More.