Cycle on Sep 7, 2026

I will start by listing the current research threads and watches associated with this topic to see if there are any specific questions, triggers, or catalysts we are tracking regarding Nvidia and the AI capex cycle.

I will pull the current market view for Nvidia (ticker NVDA) to analyze its latest financial performance, valuation, and forward outlook.

I will look up the notes to see what was previously written about Nvidia's capex story and any other details that can help me understand the context, such as the current status of the AI capex cycle, the $1.3 trillion in 2027 forecasts, and the off-balance-sheet financing models.

I will perform a web search to check for the latest news on Nvidia's capex story, its AI infrastructure buildout, demand trends, and any signs of slowdown or acceleration. Let's do a web search with `query="Nvidia AI capex demand 2026"` or similar, with `recent=true` or freshness="week" to see what the latest developments are in September 2026.

This is a fascinating set of search results. Let's analyze what's going on: 1. Nvidia reported its Q2 FY2027 (ended July 31, 2026) results recently. - Wait, let's look at the dates: "For the quarter ended July 31, 2026, the company reported $96.22 billion in revenue and a 63.7% profit margin." (This is Q2 FY2027!). - Let's check the revenue trajectory from the market view: - 2026-07-31: rev $96.22B | op $63.73B | net $59.69B | ebitda $72.63B - 2026-04-30: rev $81.61B | op $53.54B | net $58.32B | ebitda $71.00B - 2026-01-31: rev $68.13B | op $44.30B | net $42.96B | ebitda $51.28B - 2025-10-31: rev $57.01B | op $36.01B | net $31.91B | ebitda $38.75B - This shows incredible sequential growth: $57.01B -> $68.13B -> $81.61B -> $96.22B. - Let's search for "NVIDIA Q2 FY27 earnings" or "Nvidia Q2 FY2027 results transcript" to get the exact details on Nvidia's Q2 FY27 earnings call and their guidance/comments about the capex story and demand. - Also, let's look at the AWS GPU expansion and Amazon's capex: "Amazon.com, Inc. (NASDAQ:AMZN) is accelerating AI and logistics bets... boosting 2026 capex to ~$220B". "AWS will add 2M NVIDIA GPUs in 2027–28 after 1M in 2026". - Let's check the PwC report: "PwC Maps $31.6 Trillion AI Data Center Buildout Through 2050", expecting global data center capex to hit $31.6 trillion by 2050, with spending rising from $800B in 2026 to $1.8T annually. - Let's check the VanEck post: "NVIDIA Q2 FY27 Earnings Call: What It Means for SMH: Signals sustained and broadening AI infrastructure spend, with hyperscale CapEx now projected to reach nearly $800 billion in 2026 and $1.3 trillion in 2027 across the top five cloud providers. (Investing.com, 2026)" - Let's check the Milford Asset insight: "Investors very much like the uplift in revenue growth forecast for 2028 to 70% indicating the company believes the AI capex cycle has plenty of runway ahead." Let's do some more web searches to fetch these specific articles and read them in detail. Let's look for the TradingView article on AWS/Amazon capex, the PwC report, and Nvidia's Q2 FY2027 earnings details.

This is extremely rich! Let's examine this: - Amazon (AMZN) capital spending is projected to rise to roughly **$220 billion in 2026**, which is up $20 billion from its prior plan. This is a massive increase. - "AWS will add 2M NVIDIA GPUs in 2027–28 after 1M in 2026, plus NVIDIA CPU nodes, NVLink Fusion, Nemotron support on Bedrock/SageMaker, Nitro/EFA networking, and 100k GPUs for U.S. federal AI." - "Amazon says AI infrastructure spend (servers, networking) should reach breakeven in 2–3 years; management cites stronger AI demand and potential faster cloud growth and profit expansion." - "Debt markets price higher risk for Amazon (AMZN) as AI investments raise funding needs; long-term rates and credit costs now more directly affect Amazon's future borrowing costs." This means the Amazon capex story is extremely intact and in fact accelerating, with Amazon boosting 2026 capex by $20 billion to $220 billion, and AWS adding 2 million Nvidia GPUs in 2027-2028 (after 1 million in 2026). Let's check the status of the thread/watch for AWS GPU deployment: Thread watch: `AWS 2 Million Nvidia GPU Deployment (Blackwell Ultra / Rubin)` - `Monitor the successful deployment of AWS's committed 2 million additional Nvidia GPUs (Blackwell Ultra, Rubin, and Rubin Ultra) across its global cloud infrastructure by the end of 2028.` Wait, has this watch fired or is it ongoing? The watch says "by the end of 2028" so it's a future dated catalyst. But we have a major update on it! Let's record an event on it or check if there is another watch. Wait, let's look at the other watches: Is there a watch for Google or Microsoft or Meta? Let's search for "NVIDIA Q2 FY2027 earnings call transcript" or "Nvidia Q2 FY27 earnings capex" to find more details of Nvidia's recent earnings call (reported in late August 2026). Let's do a search on that.

This is an incredibly detailed transcript and set of reports. Let's analyze what we have here. First, let's look at the topic brief: "Is Nvidia's AI capex story still intact?" We have incredibly rich, fresh data from Nvidia's Q2 FY2027 earnings call, which was held on August 26, 2026. Let's summarize the key points: 1. **Nvidia's financial results are blowout:** - **Q2 FY2027 Revenue:** $96.22 billion, up 106% YoY (beating expectations of $92.07 billion). - **Q2 FY2027 Non-GAAP EPS:** $2.22, up 120% YoY (beating expectations of $2.09). - **Q2 Data Center Revenue:** $89.0 billion (92% of total), up 18% sequentially and up 106% YoY. - **Q3 FY2027 Guidance:** $108 billion (plus or minus 2%), beating street expectations of $103.8 billion. - **Preliminary FY2028 Guidance:** Nvidia gave a highly unusual full-year forward guide, expecting **FY2028 revenue to grow approximately 70% YoY**, driven by robust demand. This is a supply-constrained outlook. "We expect supply to remain a bottleneck, at least through the end of fiscal year 2028." 2. **Hyperscale Capex Numbers:** - Colette Kress (CFO) stated: "CapEx by the top 5 hyperscalers is expected to reach nearly **$800 billion in 2026** and **$1.3 trillion in 2027**." - This directly confirms that the capex numbers are not just intact but are being explicitly cited by Nvidia's management as the baseline for their forward projections. - Amazon (AWS) has committed to deploying an additional **2 million GPUs** starting in Q2 FY27 (this quarter) through Q2 FY29, along with Vera CPUs (some integrated with Rubin, others standalone). This is a massive expansion, with Amazon's 2026 capex rising to **$220 billion** (up $20 billion from prior plans). 3. **Segment Diversification (Hyperscalers vs. ACIE):** - Nvidia's data center business is split into two major segments: - **Hyperscalers:** $49 billion in Q2, up 13% sequentially. - **ACIE (AI Cloud, Industrial, and Enterprise):** $40 billion in Q2, up 25% sequentially and 138% YoY. This includes NeoClouds, sovereigns, and enterprise. - Jensen Huang and Colette Kress noted that ACIE represents about **half of their data center business** and is growing at **100% YoY**, making Nvidia's growth story much broader than just the top 5 US hyperscalers. - NeoCloud partners are expected to exit 2026 with **8 gigawatts (GW)** in total installed capacity, up from ~3 GW at the end of 2025. 4. **Nvidia's Unique Structured Financing & Ecosystem Support (The "Circular Financing" Debate):** - Nvidia addresses the "circular financing" concerns head-on. - **Frontier AI Labs (OpenAI, Anthropic):** Nvidia has invested **nearly $50 billion** in these labs. "We recognize the scale of this support, and we know some will call this circular financing. We see it differently... The technology leadership is proven and their customer traction and usage are skyrocketing... and our risk is limited. The NVIDIA compute platform is fungible and durable and can be redeployed to support other customers." - **$500B Wall Street Compute Financing Platform:** Nvidia partnered with 6 major infrastructure capital providers (Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs, and KKR) to raise over **$500 billion of third-party capital** to fund AI lab infrastructure at attractive rates. - **SoftBank Energy Partnership (PORTS-Pike Tech Campus in Ohio):** SoftBank Energy and OpenAI are deploying a **4.25 GW AI factory campus** utilizing OpenAI. Each generation of Nvidia's systems at this campus could represent **1.5 million GPUs**, with OpenAI committing to substantial deployments (approximately 12 GW of Nvidia compute) through 2030. - **NeoCloud Revenue-Sharing Model:** Instead of pure bare-metal leases or requiring massive offtake guarantees, Nvidia introduced a **revenue-sharing structure** where Nvidia provides a take-or-pay minimum revenue guarantee to lenders to underwrite NeoCloud projects, and in exchange, shares in the NeoCloud's rental revenues. "In this model, we get paid twice, once on the hardware sale and again, through the share of rental revenue..." 5. **Gross Margin Reset & Memory Headwinds:** - Nvidia is experiencing extreme pricing conditions in HBM (High Bandwidth Memory). "The magnitude of the price increase has exceeded our prior expectations and are headed even higher into next year." - As a result, Nvidia is resetting gross margin expectations: - **Q3 FY2027 Gross Margin Guidance:** 74% (plus or minus 50 bps). - **Q4 FY2027 Bottom:** Expected to bottom in the **71% to 72%** range. - **FY2028 Settling:** Expected to settle in the **72% to 73%** range, as executed price increases (over 15% server price hikes) take effect in Q1. - This gross margin reset is a crucial detail that we must capture. It's a slight downgrade from their previous high-70s margins, but they are direct about it and have a clear recovery path through price hikes starting in early 2027 (Q1 FY28). 6. **Product Ramps (Vera Rubin):** - Nvidia commenced production shipments of **Vera Rubin** in August 2026 (this month), receiving purchase orders from every major hyperscaler. They expect Vera Rubin to be the fastest product ramp in company history, accounting for about **20% of data center revenue in Q3 FY27**. - **Vera CPU:** In full production, with trailing 12-month CPU/networking revenue exceeding $5 billion for Grace, and Vera CPU revenue expected to more than double in FY2028. - **Groq 3 LPX:** In full production, setting records on token interactivity. First volume shipments later this quarter, with Nebius as the lead adopter. Let's check our threads and watches and see what we can resolve or update. First, let's look at the open watches: 1. `Nvidia Q3 FY2027 Earnings Release and Gross Margin Performance` (fire date 2026-11-17) - This is for Q3 earnings, which is scheduled for November 17, 2026. This watch is not yet fired. 2. `Nvidia Q4 FY2027 Gross Margin Bottoming (71%-72%)` (fire date 2027-02-25) - Not yet fired. 3. `Nvidia >15% Server Price Hikes Effective and Gross Margin Impact` (fire date 2027-01-15) - Not yet fired. 4. `Nvidia FY2028 Revenue Growth (Guided ~70%)` (fire date 2028-02-25) - This was just guided on this call! Colette Kress said: "Our preliminary expectation is for fiscal year '28 revenue to grow approximately 70% year-over-year...". Wait, the watch is to verify if full-year FY2028 revenue meets or exceeds the guided 70% growth rate, so it will fire in Feb 2028. 5. `Nvidia FY2028 Gross Margin Settling (72%-73%)` - Will fire in Feb 2028. 6. `AWS 2 Million Nvidia GPU Deployment (Blackwell Ultra / Rubin)` - "Monitor the successful deployment of AWS's committed 2 million additional Nvidia GPUs... by the end of 2028." This was just announced on this earnings call: "AWS is deploying an additional 2 million GPUs starting this quarter through the second quarter of fiscal '29". We can record an event or update this watch to reflect that it has been officially committed and announced! Wait, let's see if we should record an event on `AWS 2 Million Nvidia GPU Deployment (Blackwell Ultra / Rubin)`. The watch has cardinality `one_shot` and is for the *successful deployment* by the end of 2028, so it doesn't resolve yet, but we can record an ongoing event or keep it open. Wait, let's check if there are other watches we can update or resolve. Let's see: `Alphabet $40B At-the-Market (ATM) Stock Program Execution` (ongoing) - No new data yet. `SK Telecom 2GW AI Data Center First Facility Live` (one_shot, 2027-12-31) - Not yet. `SpaceX Starmind Satellite Orbital Data Center Launch` (one_shot, 2027-06-30) - Not yet. `HBM Memory Component Pricing and Supply Bottlenecks` (ongoing) - This watch is ongoing. We have a major update on this! Nvidia officially guided that HBM price increases are exceeding expectations and driving a gross margin reset to 71%-72% in Q4 FY27, before recovering to 72%-73% in FY28. We should record an event on this watch! Let's check the ID of the HBM watch: `019fbaba-2a7b-7e80-b885-9f29c3a7f0ee`. Let's record an event on it. It's ongoing, so `resolves: false`. Let's check other watches: `Nvidia $500B Wall Street Compute Financing Platform Capital Deployment` (ongoing, ID `019fee38-1dbd-73a3-8fa2-113b552feff5`). Wait! On the earnings call, Colette Kress announced: "With these partnerships, building on our unique fungible and durable computing platform, the AI labs will be able to build and assess AI infrastructure funded by long-term institutional capital at relatively attractive rates." And: "Last week, we announced that we secured land power shell capacity through our partnership with SoftBank Energy to exclusively host NVIDIA compute at their Portsmouth Campus. The initial deployment expected to support 4.25 gigawatts of AI factory capacity will be utilized by OpenAI." This is a huge milestone for the $500B platform and the OpenAI deal! Let's record an event on `Nvidia $500B Wall Street Compute Financing Platform Capital Deployment` (ID `019fee38-1dbd-73a3-8fa2-113b552feff5`). Since it's ongoing, we can record an event and keep it open (`resolves: false`). Let's check the OpenAI PORTS-Pike Tech Campus watch: `OpenAI Ohio PORTS-Pike Technology Campus Phase 1 Live` (one_shot, ID `01a0364e-15f1-702b-8163-2daf6b8543fb`). This watch is for when it goes live. So it doesn't resolve yet, but we have the official announcement of the 4.25 GW campus deployment for OpenAI. Let's check if there are other watches that we can record events for. Let's write down the findings first. We need to write findings that update our existing notes. Let's look at the existing notes: 1. `big-tech-capex-reaches-710b-in-2026` -> Let's update this! - Title was: "Big Tech's AI Capex Surpasses $800 Billion in 2026 on Track for $1.3 Trillion in 2027 as Nvidia Projects 70% Growth" - Wait, the slug is `big-tech-capex-reaches-710b-in-2026`. We can reuse this slug to update it with the latest Q2 FY27 numbers, Amazon's $220B capex increase, and AWS's 2 million GPU commitment. 2. `off-balance-sheet-ai-infrastructure-financing-models` -> Let's update this! - This note is about the off-balance-sheet JVs, the $500B Wall Street platform, the $105B OpenAI deal, SoftBank Energy, and the Portsmouth campus. - We have exact quotes and numbers from the Q2 FY27 earnings call: Nvidia's $50B investment in frontier labs, the SoftBank Energy Portsmouth campus (4.25 GW, supporting 1.5 million GPUs per generation, OpenAI committing to 12 GW of Nvidia compute through 2030), and the NeoCloud revenue-sharing model. This is an incredible update! 3. `nvidia-server-price-hikes-and-margin-protection` -> Let's update this! - This note is about HBM cost inflation and Nvidia's gross margin reset. - We have the exact gross margin guidance from the Q2 call: 74% in Q3, bottoming in Q4 at 71%-72%, settling in FY28 at 72%-73% as price increases take effect. Let's double-check if there are other notes we should update or create. Wait, let's look at the instructions: "ONE note per theme, never a combined write-up. A finding TELLS THE STORY THROUGH CONCRETE EVIDENCE... carried 1-2 REPRESENTATIVE VERBATIM QUOTES, copied exactly, each kept with the link it came from... Slugs are stable identities... Before you write, scan your prior findings for the slug that already owns this subject and reuse it". Let's verify the exact details of the three notes we want to update: 1. `big-tech-capex-reaches-710b-in-2026` 2. `off-balance-sheet-ai-infrastructure-financing-models` 3. `nvidia-server-price-hikes-and-margin-protection` Let's review the exact quotes we will use from the Q2 FY27 earnings transcript: For `big-tech-capex-reaches-710b-in-2026`: - Quote: "CapEx by the top 5 hyperscalers is expected to reach nearly $800 billion in 2026 and $1.3 trillion in 2027." - Quote: "AWS is deploying an additional 2 million GPUs starting this quarter through the second quarter of fiscal '29 along with Vera, CPUs, some integrated with Rubin, other standalone." - Quote: "Our preliminary expectation is for fiscal year '28 revenue to grow approximately 70% year-over-year, although we will work to close the supply-demand gap, we expect supply to remain a bottleneck, at least through the end of fiscal year '28." - Link: `https://www.fool.com/earnings/call-transcripts/2026/08/31/nvidia-nvda-q2-2027-earnings-call-transcript/` For `off-balance-sheet-ai-infrastructure-financing-models`: - Quote: "First, we've invested nearly $50 billion in the Frontier AI labs... Further, to support the Frontier Labs infrastructure build-outs, we recently announced partnerships with 6 of the world's leading infrastructure capital providers, Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR, to establish financing platforms that will raise over $500 billion of third-party capital." - Quote: "Last week, we announced that we secured land power shell capacity through our partnership with SoftBank Energy to exclusively host NVIDIA compute at their Portsmouth Campus. The initial deployment expected to support 4.25 gigawatts of AI factory capacity will be utilized by OpenAI." - Quote: "Rather than allocating their entire capacity to a single long-term offtake guarantee that lenders typically require to finance a data center independently, we have introduced a revenue-sharing structure. NVIDIA provides a take-or-pay commitment on a portion of the facility's capacity, a minimum revenue guarantee... In this model, we get paid twice, once on the hardware sale and again, through the share of rental revenue..." - Link: `https://www.fool.com/earnings/call-transcripts/2026/08/31/nvidia-nvda-q2-2027-earnings-call-transcript/` For `nvidia-server-price-hikes-and-margin-protection`: - Quote: "We are experiencing extreme pricing conditions in memory. The magnitude of the price increase has exceeded our prior expectations and are headed even higher into next year... For Q3, we expect GAAP and non-GAAP gross margins to be 74% plus or minus 50 basis points. We expect margins to bottom in Q4 in the 71% to 72% range before settling at 72% to 73% in fiscal year '28, as executed price increases take effect in Q1." - Link: `https://www.fool.com/earnings/call-transcripts/2026/08/31/nvidia-nvda-q2-2027-earnings-call-transcript/` Let's verify if there are any other notes we should write. No, these three cover the main themes beautifully. Let's call the `record_event` tool first to record the events on the watches that have fired or have major updates. Wait, let's check the watches again. - `HBM Memory Component Pricing and Supply Bottlenecks` (

I will record events on the active watches that have received major updates during Nvidia's Q2 FY2027 earnings call. First, I'll update the watch tracking HBM memory component pricing and supply bottlenecks.

I will now record an event on the watch tracking the launch and deployment of Nvidia's $500 billion partnership with Wall Street infrastructure capital providers, as major milestones were officially detailed on the earnings call.

I will now write the updated finding for the hyperscaler capex trajectory under the slug `big-tech-capex-reaches-710b-in-2026`. This update captures Nvidia's Q2 FY2027 blowout earnings, the formal projection of $1.3 trillion in hyperscaler capex for 2027, Amazon's capex increase to $220 billion, and AWS's massive 2 million GPU commitment.

I will now write the updated finding for the structured financing models under the slug `off-balance-sheet-ai-infrastructure-financing-models`. This captures Nvidia's $50B direct investment in frontier labs, the $500B Wall Street compute financing platform, the 4.25 GW SoftBank Energy Portsmouth campus deployment for OpenAI, and the new NeoCloud revenue-sharing model.

I will now write the updated finding for Nvidia's margin protection and gross margin reset under the slug `nvidia-server-price-hikes-and-margin-protection`. This captures the extreme HBM memory pricing headwinds, the guided gross margin reset (bottoming at 71%-72% in Q4 FY27), and the recovery path via >15% server price hikes taking effect in early 2027.

I will now complete the research cycle by calling the `finish_cycle` tool. We have updated three major findings and recorded events on our active watches based on Nvidia's Q2 FY2027 earnings call.

I have completed this research cycle on Nvidia's AI capex story. The findings are highly concrete, detailed, and directly grounded in the authoritative public-markets data and the transcript of Nvidia's Q2 FY2027 earnings call held on August 26, 2026. ### Executive Summary: Is Nvidia's AI Capex Story Still Intact? Yes, Nvidia's AI capex story is not only intact but is **accelerating and broadening** across multiple vectors. The primary constraint on the AI boom remains **supply and physical infrastructure (power/land)** rather than a lack of customer demand or return on investment. Here are the key takeaways from the research: 1. **Massive Hyperscale Capex Expansion:** - Nvidia's CFO officially projected that capital expenditures across the top five hyperscalers will reach nearly **$800 billion in 2026** and balloon to **$1.3 trillion in 2027**. - **Amazon (AWS)** has dramatically accelerated its buildout, boosting its 2026 capex budget to **$220 billion** (up $20 billion) and committing to deploy an additional **2 million Nvidia GPUs** (along with Vera CPUs and Rubin GPUs) starting this quarter through Q2 FY2029. - Nvidia reported record **Q2 FY2027 revenue of $96.22 billion** (+106% YoY) and guided to **$108 billion in Q3 FY2027**. - Nvidia took the highly unusual step of guiding to a full-year forward outlook, projecting **approximately 70% revenue growth in FY2028**. This is a strictly supply-constrained projection, with supply expected to remain a bottleneck through the end of FY28. 2. **Broadening Beyond Hyperscalers (The ACIE Segment):** - Nvidia’s AI Cloud, Industrial, and Enterprise (ACIE) segment grew **138% YoY to $40 billion** in Q2, representing roughly half of their data center business. - Specialized GPU clouds ("neoclouds") are expanding rapidly, expected to exit 2026 with **8 gigawatts (GW)** of installed capacity, up from ~3 GW at the end of 2025. 3. **Structured Off-Balance-Sheet Financing & The "Circular Financing" Debate:** - Nvidia is actively engineering its own demand by structuring off-balance-sheet joint ventures to bridge the capital gap for Frontier AI labs. - **$500B Wall Street Platform:** Nvidia partnered with Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs, and KKR to raise over **$500 billion in institutional capital** to fund AI lab infrastructure. - **OpenAI Portsmouth Campus:** Nvidia and SoftBank Energy secured a **4.25 GW AI factory campus** in Ohio (PORTS-Pike) to exclusively host OpenAI workloads. Each hardware generation at this site represents **1.5 million GPUs**, with OpenAI committing to **12 GW of Nvidia compute** through 2030. - **NeoCloud Revenue-Share:** Nvidia introduced a "take-or-pay" minimum revenue guarantee to lenders to underwrite NeoCloud projects in exchange for a share of their rental revenues above the floor. This allows Nvidia to get paid twice—once on the hardware sale, and again via high-margin recurring rental streams. 4. **Gross Margin Reset & Memory Headwinds:** - Nvidia is experiencing extreme pricing conditions in HBM (High Bandwidth Memory) that have exceeded expectations. - Consequently, Nvidia has reset gross margin expectations: guiding to **74% in Q3**, bottoming in the **71%-72% range in Q4 FY27**, and settling back to **72%-73% in FY2028**. - To protect its margins, Nvidia is implementing a **>15% price hike on server systems** starting in Q1 FY2028 (early calendar 2027) and is working to qualify Samsung's custom 8-layer HBM4E to ease supply bottlenecks. ### Updated Findings (Living Notes) The following notes have been updated and are persisted under the topic: - **`big-tech-capex-reaches-710b-in-2026`**: Updated with the blowout Q2 FY27 results, the $1.3T 2027 capex projection, and Amazon's $220B capex / 2M GPU commitment. - **`off-balance-sheet-ai-infrastructure-financing-models`**: Updated with details of the $500B Wall Street financing platform, the 4.25 GW SoftBank Energy PORTS-Pike campus for OpenAI, and the NeoCloud revenue-sharing model. - **`nvidia-server-price-hikes-and-margin-protection`**: Updated with the gross margin reset to 71%-72% in Q4 FY27 due to HBM pricing headwinds and the >15% price hike recovery strategy. ### Fired/Updated Watches - **`HBM Memory Component Pricing and Supply Bottlenecks`** (ID: `019fbaba-2a7b-7e80-b885-9f29c3a7f0ee`): Recorded event regarding Nvidia's guided gross margin reset due to HBM pricing. - **`Nvidia $500B Wall Street Compute Financing Platform Capital Deployment`** (ID: `019fee38-1dbd-73a3-8fa2-113b552feff5`): Recorded event regarding the formal details of the $500B platform and the 4.25 GW PORTS-Pike campus deployment.