Born to Give Them Second Birth | | by the Gilder Team
12/25/2024 | | | SPONSORED CONTENT These 7 Stocks Will Be Magnificent in 2025 The original Magnificent Seven stocks are true market leaders. However, as these companies have matured, the opportunities to replicate their meteoric returns have become fewer. Where can you find the next big winners? We have identified what we believe will be the Next Magnificent 7. And you can get all 7 names absolutely FREE. These global companies are positioned to surge in 2025.
Get Your Copy of "These 7 Stocks Will Be Magnificent in 2025" Here. | | | At just about noon on Christmas Eve, the parts of our brains devoted to technology and investment become oddly slow to respond. Like the shepherds, we become attuned to events more important than work. We are inclined not to toil but to celebration of the coming of the One from whom come all our bounties and all the fruits of our labors.
With our minds and hearts elsewhere, we offer for today's Guideposts a "best of," a recent collaboration between George and our newest team member Dr. Robert Castellano.
From all of us at the Gilder Team, have the merriest of Christmases and a bountiful new year.
George Gilder
John Schroeter
Steve Waite
Richard Vigilante
Robert Castellano
***************************
As we escape a tempestuous political season, we might imagine that Trump's victory somehow affirmed his claims of an economic debacle bruising America's middle class, with China rooking and reaming our intellectual property as a prime culprit.
Yet we semiconductor guys, with our heads in the silicon—for the last 50 years or so—find it hard to see our economy as crippled. Silicon remains king, and the United States remains king of silicon, with our corporations commanding, according to various estimates, between 49% and 70% of global market cap, compared to Chinese corporations at around 6%. Nearly all that U.S. bulge feeds on chips. Meanwhile, we are pioneering a new paradigm of artificial intelligence (AI) and it's nearly all still based on silicon.
Beyond politics, two giants, NVIDIA (NASDAQ: NVDA) led by Jensen Huang, in his signature swashbuckling black leather, and Intel (NASDAQ: INTC) ruled by Pat Gelsinger, in a sleek grey blazer and a rakish tee-shirt, wrangle publicly over the fate of Moore's Law: the biennial doubling of computer power on chips. It was ordained by Intel co-founder Gordon Moore with the help of his advisor, and ours, Carver Mead, who did key research and named the law way back in 1965. Moore's Law epitomizes Intel.
Jensen says "it's over," debunked and displaced by Jensen's Law: "NVIDIA is achieving millionfold advances every decade." Pointing to the continued ascent of transistor numbers on Intel chips, Pat Gelsinger says, "Moore's Law is alive and well!"
With Nvidia shares now worth some $3.5 trillion, making it the world's most valuable company, and Intel market cap some 35 times lower, Jensen's law seems to rule. But Robert Castellano, our new colleague, tells us that they are both right. They just need to stop competing in the wrong directions and adopt win-win strategies.
The semiconductor industry's relentless growth in AI datacenter computing has exposed a critical bottleneck: manufacturing capacity. Nvidia, the leader in AI processors, massively relies on Taiwan Semiconductor Manufacturing Company (TSMC) for leading-edge production. However, TSMC's leading edge wafer fabs, operating with geometries in the billionths of a meter at 3 nanometers (nm) and 5nm, are fully booked. Beyond Nvidia, TSMC's customer list is a constellation of such mostly fab-less leaders as Apple (NASDAQ: AAPL), Advanced Micro Devices (NASDAQ: AMD), Arm Holdings PLC (NASDAQ: ARM), Qualcomm Inc. (NASDAQ: QCOM), MediaTek Inc. (OTC: MDTTF) (TW: 2454), and Broadcom Inc. (NASDAQ: AVGO) that provide silicon designs for modern electronics and AI.
Also on the list, surprisingly, is Intel, which runs vast wafer fabs around the globe. But Intel, unlike TSMC, also designs scores of leading-edge microprocessors that its own fabs cannot currently handle. Intel has outsourced production of some of its nanometer-node chips to TSMC because of delays in Intel's internal manufacturing.
This dual reliance on TSMC by Nvidia and Intel strains an already overloaded production pipeline. It risks delays in delivering Nvidia's vital AI processors. Without Nvidia's leading-edge chips, companies such as Apple, Meta (NASDAQ: META), Amazon.com Inc. (NASDAQ: AMZN), and Alphabet (NASDAQ: GOOGL) that account for most of that huge U.S. dominance in global market cap, would flounder.
A strategic realignment, where Intel Foundry Services (IFS) absorbs Intel's non-AI production from TSMC, could unlock significant capacity for Nvidia and reshape the competitive dynamics of the semiconductor industry. | | Have You Seen This $11 Trillion 'Tech Strip?' While many folks today are wondering what to do with their money… a revolutionary "sheet" of new technology has quietly sparked an $11 trillion tech revolution.
Investors who get in FIRST have a rare chance to position themselves in front of a tsunami of profits.
Click here to see how anyone can profit fast. | | | Realignment: Leveraging Intel Foundries to Free TSMC Capacity
Under this proposed realignment, Intel Foundry Services would take over the production of non-AI workloads currently manufactured at TSMC, allowing the Taiwanese foundry to allocate more of its 3nm and 5nm capacity to Nvidia's AI processors. This shift would include products such as Intel's Meteor Lake "chiplets," "ARC" gaming graphics processors (GPUs), and integrated graphics solutions. These chips require less critical timelines compared to Nvidia's industry-leading and enabling Hopper and Blackwell GPUs. | Table 1: Intel's TSMC Workload and Strategic Shift to IFS | | Category | TSMC Workload for Intel | Shift to IFS | Impact | | CPUs | Chiplets for Meteor Lake, Raptor Lake | Intel 20A/18A at IFS | Frees TSMC capacity for Nvidia's AI chips. | | GPUs | Intel ARC discrete GPUs | Intel 3 at IFS | Enables TSMC to prioritize Nvidia's Hopper and Blackwell. | | Chipsets | Integrated graphics and IoT solutions | Mature nodes at IFS | Optimizes TSMC's resources for advanced nodes. | | Source: The Information Network |
This redistribution not only addresses Nvidia's immediate supply chain challenges but also positions Intel as a key partner in Nvidia's future success. Additionally, TSMC would benefit from increasing its high-margin AI business, while Nvidia would gain more production capacity for its flagship GPUs.
Financial Implications: NVDA, TSM, and INTC
The proposed realignment between Nvidia, Intel, and TSMC presents a strong business case for all three stakeholders. Nvidia stands to benefit the most, as additional capacity at TSMC would alleviate supply constraints for its high-margin AI processors. The shift would enable Nvidia to ramp production of Hopper and Blackwell GPUs, driving significant revenue growth to meet surging demand.
For TSMC, reallocating its capacity to Nvidia's advanced AI chips would increase profitability, as AI processors command higher average selling prices (ASPs) compared to consumer and non-AI workloads. Meanwhile, Intel Foundry Services (IFS) would gain from the influx of new business, including its own internal products and potentially Nvidia's less advanced products, helping Intel reestablish itself as a credible foundry player. | Table 2: Financial Benefits by Stakeholder | | Company | Estimated Revenue Impact ($B) | Details | | Nvidia | +$5–8 Billion | Increased production of Hopper and Blackwell GPUs to meet unmet demand. | | TSMC | +$2–4 Billion | Higher ASPs from reallocating capacity to Nvidia's high-margin AI processors. | | Intel (IFS) | +$4–6 Billion | New foundry customer business from Intel's own product shift and potential Nvidia collaborations. | | Source: The Information Network |
This strategic shift could enable Nvidia to further secure its leadership in AI while providing Intel and TSMC with significant new revenue streams. This alignment positions all parties to address the immediate bottlenecks in the semiconductor supply chain while capitalizing on the expanding demand for AI computing.
The main overlap between Intel and Nvidia's product lines is GPUs and data center hardware, but their focus differs. Nvidia dominates in high-performance AI and data center GPUs, while Intel focuses on consumer GPUs, such as ARC and integrated graphics solutions. This divergence creates opportunities for strategic collaboration without major competitive conflicts. Shifting Intel's non-AI production from TSMC to Intel Foundries would further align their strengths, enabling TSMC to focus on Nvidia's high-demand AI processors while Intel expands its gaming GPU and chipset businesses. | Table 3: Product Overlap and Strategic Fit | | Product Line | Intel | Nvidia | Overlap & Strategic Fit | | Gaming GPUs | ARC GPUs (low current share) | GeForce RTX (dominant) | Intel could grow in gaming, complementing Nvidia's AI focus. | | AI Processors | Gaudi, Habana (limited) | Hopper, Blackwell | Nvidia dominates; Intel can enhance foundry collaboration. | | Data Center GPUs | Xe GPUs (limited adoption) | Hopper, Blackwell | Nvidia leads, with Intel trailing in adoption and revenue. | | Source: The Information Network | | | Where are the markets going? Do you know where the S&P 500 is going?
Well… these urgent A.I. flash signals are shocking us.
Learn what they mean – LIVE. | | | Intel's mid-range consumer GPU focus complements Nvidia's high-end GPU leadership.
This synergy reduces direct competition and makes the proposed collaboration more feasible, benefiting both companies by aligning resources and priorities.
Competition in Gaming and Graphics
What opportunities would Intel be forgoing by yielding TSMC capacity? Practically speaking, none, if it could successfully manufacture its own non-AI chips. It has the equipment and fabs to do that. If yields are a bit low to start, which is why this work ended up at TSMC in the first place, there is no practical alternative to getting on the learning curve and getting better. The switchover could be done gradually as yields improve, but it must be done ultimately or IFS has no future.
In theory, it might seem that INTC, by releasing some of TMSC's most advanced process lines to NVDA, it would be yielding the AI data center market without a fight. In practice, that has already happened. Nvidia's lead in that market already is overwhelming compared to both INTC and AMD.
As the graph below shows, INTC has been losing ground in data centers and AMD has gained—a little. Meanwhile, NVDA has transformed the market with its AI processors such as the Hopper H200 and the anticipated Blackwell GPU. NVDA's grip will only tighten in the near term. Analysts predict Nvidia's data center revenue will increase another 9% quarter-over-quarter, with a staggering 97% year-over-year growth in the upcoming quarter.
This proposed realignment of semiconductor manufacturing capacity offers a pragmatic solution to the industry's supply chain challenges with three big winners (not counting the global economy, which needs these chips)!
With Intel taking non-AI workloads back to its own foundries, Nvidia would gain additional capacity at TSMC for Hopper and Blackwell. Intel, through its foundry services, could establish itself as a critical partner for Nvidia while addressing its own revenue challenges. For TSMC, this strategy would mean higher margins and more efficient resource allocation.
Is there a moral to this story beyond the fate of three companies? Sure. To maintain U.S. global leadership in technology we need to maintain the well-knit global ecosystem that got us here and return to the win-win economics that opens the world to the future.
Sincerely,
George Gilder, Richard Vigilante, Steve Waite, and John Schroeter
Editors, Gilder's Guideposts, Technology Report, Technology Report Pro, Moonshots, and Private Reserve | | About George Gilder:
 George Gilder is the most knowledgeable man in America when it comes to the future of technology and its impact on our lives. He’s an established investor, bestselling author, and economist with an uncanny ability to foresee how new breakthroughs will play out, years in advance. George and his team are the editors of Gilder Technology Report, Gilder Technology Report Pro, Moonshots and Private Reserve. | | | | | |
Visit our YouTube Channel - Eagle Investing Network
No comments:
Post a Comment