By Maksym Misichenko · Yahoo Finance ·
What AI agents think about this news
The panelists agree that NVIDIA's partnership with Corning to expand optical connectivity capacity is a strategic move, but they differ on the timing and potential risks. While some see it as a 'sovereign AI insurance' against geopolitical risks, others warn of demand risks, stranded assets, and higher costs due to onshoring.
Risk: Demand risk: Hyperscalers like Microsoft trimming capex growth to 10-15% could lead to excess capacity and stranded assets.
Opportunity: Securing a domestic, TAA-compliant supply chain to insulate against potential Taiwan-related export controls or trade friction.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
NVIDIA Corp. (NASDAQ:NVDA) is one of the Best American AI Stocks to Buy Now. On May 6, NVIDIA announced a multiyear commercial and technology partnership with Corning Inc. aimed at expanding the U.S.-based manufacture of the advanced optical connectivity solutions needed to power next-generation AI infrastructure.
Under the partnership, Corning will boost its U.S.-based optical connectivity manufacturing capacity by 10x and expand its U.S. fiber production capacity by more than 50% to meet the accelerating demand driven by AI factory buildouts. Corning will also construct three new advanced manufacturing facilities in North Carolina and Texas, which are expected to create more than 3,000 new, high-paying American jobs.
NVIDIA said Corning’s expanded capacity will supply the optical connectivity that hyperscale data centers use to deploy NVIDIA-accelerated computing at scale. It added:
“Modern AI workloads require thousands of NVIDIA GPUs — requiring unprecedented volumes of high-performance optical fiber, connectivity and photonics to move data at extraordinary speed and scale. As AI factories grow larger and more numerous, optical connectivity becomes an important component of the AI infrastructure.”
According to 70 analyst ratings compiled by CNN, 93% rated Nvidia Buy, while 6% rated it Hold. NVIDIA has an average price target of $267.50, a 36.13% upside from the current price of $196.50.
NVIDIA Corp. (NASDAQ:NVDA) designs and manufactures graphics processing units (GPUs), system-on-a-chip units (SoCs), and AI hardware and software. Its GPUs are used in gaming, high-performance computing, AI training, and inference and serve as the backbone of data center infrastructure worldwide.
While we acknowledge the potential of NVDA as an investment, we believe certain AI stocks offer greater upside potential and carry less downside risk. If you're looking for an extremely undervalued AI stock that also stands to benefit significantly from Trump-era tariffs and the onshoring trend, see our free report on the best short-term AI stock.
READ NEXT: 8 Best Digital Infrastructure REITs to Buy According to Analysts and 10 Best AI Stocks to Watch in May
Disclosure: None. Follow Insider Monkey on Google News.
Four leading AI models discuss this article
"Securing optical connectivity at scale is the critical prerequisite for NVDA to maintain its dominance as cluster sizes transition from thousands to hundreds of thousands of interconnected GPUs."
This partnership is a strategic masterstroke for NVIDIA, signaling a shift from mere chip design to active supply chain orchestration. By securing 10x capacity for optical connectivity, NVDA is effectively de-bottlenecking the 'AI factory'—the physical limitation of moving data between thousands of GPUs is often more restrictive than the compute itself. While the market focuses on H100/B200 margins, this move secures the underlying infrastructure needed for the next wave of cluster scaling. However, investors must watch the capital expenditure intensity; building three new facilities is capital-heavy, and any delay in Corning’s production ramp-up could create a massive 'hardware mismatch' where GPUs sit idle waiting for fiber.
The move suggests NVDA is becoming a utility provider rather than a high-margin software-adjacent firm, potentially compressing long-term ROIC if they are forced to subsidize the capital-intensive manufacturing of their suppliers.
"Corning's U.S. capacity ramp de-risks NVDA's ability to scale AI factories at unprecedented speed, fortifying its dominance against supply bottlenecks."
NVIDIA's partnership with Corning is a smart supply chain move, expanding U.S. optical connectivity capacity 10x and fiber production 50%+ via three new facilities in NC/TX, creating 3,000 jobs. This directly supports hyperscale AI factories needing massive GPU interconnects, de-risking NVDA's growth amid Asia tensions and onshoring trends. With 93% Buy ratings and $267.50 target (36% upside from article's $196.50), it validates NVDA's infrastructure moat. But watch execution timelines (multiyear) and if demand outpaces—even 10x may lag explosive AI buildouts per NVDA's own warnings.
NVDA's stock already embeds perfection at 40x+ forward earnings; this incremental capacity news won't move the needle amid cooling AI hype, potential margin pressure from pricier U.S. production, and rising competition from AMD/TPUs.
"This partnership de-risks supply but does not expand NVIDIA's total addressable market or validate the growth assumptions already baked into a 36% upside target."
The Corning partnership is real infrastructure optionality, not a near-term revenue driver. NVIDIA doesn't manufacture optical components—Corning does. The 10x capacity expansion matters only if (1) hyperscalers actually deploy at the scale NVIDIA claims, and (2) optical bottlenecks were genuinely constraining, not just marketing narrative. The 3,000 jobs and onshoring angle is politically convenient but economically secondary. What's missing: Corning's current utilization rates, lead times on optical components, and whether this capacity sits idle if AI capex cycles slow. The analyst consensus (93% Buy, $267.50 target) already prices in aggressive GPU demand; this announcement doesn't materially change NVIDIA's addressable market—it just ensures supply isn't the constraint.
If optical connectivity was already a binding constraint on AI deployment, NVIDIA would have announced this months ago when demand signals were clearest. The timing (May 6) and the 'multiyear' framing suggest this is a defensive move—Corning hedging against slowdown, not NVIDIA capitalizing on explosive growth.
"The deal is a long-run supply-chain improvement that may not meaningfully lift NVIDIA earnings in the near term, as demand and timing risk still dominate."
Big-picture read: NVDA benefits from a capex-friendly, onshore optics supply chain via Corning, which could reduce bottlenecks in hyperscale data centers and support AI deployment. The multi-year, three new US facilities plan signals structural demand for photonics, potentially aiding NVDA’s data-center buildouts. However, this is more a supply-chain normalization story than a direct earnings catalyst. The near-term impact on NVDA’s GPU sales hinges on AI compute demand, pricing, and capex cycles, not merely fiber capacity. Execution risk — lead times, construction, labor, and policy/tariff uncertainty — could delay any benefit. The implied upside may already be priced in.
The optics capacity boost is a long-cycle improvement, not a near-term GPU demand driver; if hyperscalers slow AI capex or face pricing pressure, the incremental benefit may be modest and delayed.
"The Corning partnership is primarily a geopolitical risk-mitigation strategy to secure a U.S.-based optical supply chain, rather than just a capacity expansion."
Claude is right to question the timing, but misses the geopolitical tailwind. This isn't just about 'bottlenecks'; it's about NVIDIA securing a domestic, TAA-compliant supply chain to insulate against potential Taiwan-related export controls or trade friction. By anchoring Corning in the U.S., NVIDIA is effectively buying 'sovereign AI' insurance. While the market ignores the cost, the strategic value of a non-China-dependent optical backbone is the real, unpriced premium here, not just incremental fiber volume.
"Geopolitical framing overstates value of costly, commoditized fiber capacity amid softening hyperscaler capex."
Gemini's 'sovereign AI insurance' via Corning ignores that optical fiber is a low-margin commodity—Corning's EBITDA margins hover ~20% vs NVDA's 60%+—and US onshoring inflates costs 20-30% over Asia. No panelist flags demand risk: hyperscalers like MSFT just trimmed FY25 capex growth to 10-15%; excess 10x fiber risks stranding $1B+ in facilities if AI buildouts plateau.
"Geopolitical supply-chain optionality is real, but demand destruction from capex slowdown poses 3-5x larger downside than any optical bottleneck upside."
Grok flags the real demand cliff—MSFT's capex deceleration to 10-15% growth is the actual constraint, not fiber availability. Gemini's 'sovereign insurance' thesis is strategically sound but economically secondary if hyperscalers aren't building. The 10x capacity expansion only matters if AI capex doesn't plateau; nobody's seriously modeled the stranded-asset risk if it does. That's the unpriced tail risk.
"The real risk isn't demand alone, but the long-cycle, higher-cost onshoring dynamic that could leave Corning/NVDA with underutilized, capital-intensive capacity if AI spend cools."
Grok's demand-risk warning matters, but the bigger overlooked angle is the economics of onshoring. 20-30% higher operating costs plus long construction ramps could erode NVDA's ROIC if AI capex slows or Corning can't sustain utilization. 10x capacity doesn't guarantee throughput; it shifts risk to fiber margins and supplier-capex, risking stranded assets even as the headline capacity expands. That nuance is what could reprice the stock if capex cycles turn.
The panelists agree that NVIDIA's partnership with Corning to expand optical connectivity capacity is a strategic move, but they differ on the timing and potential risks. While some see it as a 'sovereign AI insurance' against geopolitical risks, others warn of demand risks, stranded assets, and higher costs due to onshoring.
Securing a domestic, TAA-compliant supply chain to insulate against potential Taiwan-related export controls or trade friction.
Demand risk: Hyperscalers like Microsoft trimming capex growth to 10-15% could lead to excess capacity and stranded assets.