THE APEX TIMES
Google deal with SpaceX set to reach about $30 billion for AI compute, SEC filing shows
Alphabet’s Google has agreed to pay SpaceX $920 million per month for access to roughly 110,000 NVIDIA GPUs and related compute capacity, according to an SEC filing tied to SpaceX’s IPO process.
Google’s cloud unit is paying SpaceX $920 million per month for AI compute capacity, a deal SpaceX disclosed in an SEC filing. The agreement runs from October 2026 through June 2029 and is valued at roughly $30 billion over its term, according to the filing.
The filing says SpaceX entered into a Cloud Service Agreement with Google on June 5, 2026. Under the contract, SpaceX will provide access to approximately 110,000 NVIDIA GPUs along with CPUs, memory, and other related components, with capacity ramping up through September for a reduced fee.
SpaceX also laid out what happens if the compute delivery misses the agreed timeline. If SpaceX fails to deliver access to the committed GPU amount by September 30, 2026, then after a one-month grace period, Google can terminate the agreement or accept fewer GPUs with a proportional reduction in the monthly fees. After December 31, 2026, either party can end the arrangement with 90 days’ notice.
The terms suggest Google is buying “bridge capacity” rather than permanently outsourcing its infrastructure. In a statement reported by TechCrunch, a Google representative described the deal as timely support tied to demand for Google’s AI products, including Gemini Enterprise, as Google scales capacity for customers.
The deal underscores how compute has become a limiting factor for AI development and deployment. In an investor presentation posted to the Alphabet investor blog on June 3, 2026, Alphabet’s CEO Sundar Pichai said demand for AI solutions and services is exceeding available supply, adding that supporting this at scale “requires massive compute investments.”
Alphabet’s presentation also put numbers on the pressure. Pichai said Alphabet spent about $31 billion on capital expenditures in 2022, expected 2026 spending to be about $180 to $190 billion, and indicated that much of the increase would go to technical infrastructure. The SpaceX compute agreement is consistent with that backdrop, as Alphabet seeks additional capacity while it expands data-center buildout.
For SpaceX, the agreement adds a large, contract-backed revenue stream that arrives as it prepares for public-market scrutiny. TechCrunch reported that SpaceX did not specify which data center Google would use for the compute under the contract, leaving open questions about where the capacity sits and how it will be managed operationally.
Why It Matters
- The size and duration of the compute contract highlight how AI infrastructure procurement is increasingly driven by capacity scarcity, not just pricing.
- The deal provides SpaceX with a predictable, multi-year revenue line that could matter to investor perception as the company moves toward an IPO.
- Google’s buy-versus-build choice may shape near-term margins and capex timing, since third-party compute can act as a buffer while data-center capacity expands.
- The agreement’s delivery and termination clauses put measurable performance expectations on SpaceX, and future disclosures could clarify whether those targets are met.
- Which physical data centers are used, and how the compute is delivered during ramp-up, remain key operational details that were not specified in secondary reporting.
Sources
Key Facts
- SpaceX disclosed a Cloud Service Agreement with Google LLC entered on June 5, 2026, in an SEC filing tied to SpaceX’s IPO process.
- The agreement provides access to approximately 110,000 NVIDIA GPUs plus CPUs, memory, and related components.
- Google will pay SpaceX $920 million per month from October 2026 through June 2029, with capacity ramping through September at a reduced fee.
- If committed GPU capacity is not delivered by September 30, 2026, Google can terminate after a one-month grace period or accept fewer GPUs with a pro-rata reduction in monthly fees.
- After December 31, 2026, either party can terminate the agreement with 90 days’ notice.
- A related report published on June 7, 2026 highlighted the same arrangement and said the monthly total amounts to about $30 billion over the term, though the headline used a different dollar figure than the SEC-described monthly payment.
Technology Related
Jensen Huang’s “Buy at a Discount” remark returns to focus as Nvidia shares rise and an AI basket gains
A CEO message to investors in June has been replayed after Nvidia’s stock moved higher over the following months, alongside gains in a broader AI peer group. Analysts caution that short-term trading often reflects many forces beyond a single CEO comment.
AMD says it is expanding its AI infrastructure footprint in Saudi Arabia
The chip designer announced a new platform initiative in Saudi Arabia, while investors appeared focused on how quickly the move could translate into additional AI-related revenue. AMD shares were little changed in Monday premarket trading.
Nvidia shares show a rare trading pattern, underscoring how investors are rethinking semiconductor correlations
A market-linked read of Nvidia’s stock behavior suggests its relationship with broader semiconductor moves has shifted, a change that can affect hedging, positioning, and how traders interpret near-term momentum.
Nvidia backs MediaTek with $3.5 billion convertible-bond deal, indicating a push for local AI
Nvidia is investing $3.5 billion in Taiwan-based MediaTek via convertible bonds, deepening an existing AI partnership. The move points to growing interest in deploying AI closer to devices, not just in data centers.
FTC and 22 states sue Amazon, alleging it manipulated online ad auctions
Regulators claim Amazon’s advertising technology inflated costs for advertisers, saying the alleged conduct led to more than $20 billion in overcharges for about 1.2 million advertisers.
Alphabet’s Google says Gemini-powered “Teamwork” agents solved open math, built a CPU simulator, and improved core open-source libraries
In an update to its Antigravity multi-agent framework, Google reports results spanning theoretical computer science benchmarks, cycle-accurate hardware emulation, and upstream performance contributions to widely used software libraries.
Nvidia hardware momentum meets a new choke point: copper, not cash, HIVE Digital’s Frank Holmes says
A Wall Street executive argues that today’s AI funding is not the limiting factor. The bottleneck, he says, is the physical supply chain behind data centers, where power and copper wiring needs can outstrip available materials.
Broadcom’s Sept. 2 earnings set up a high-stakes test for its AI narrative
Ahead of its next quarterly report, Broadcom is drawing attention from investors who are trying to separate short-term uncertainty from longer-term demand linked to artificial intelligence.
Palantir CEO Alex Karp pushes back on “tokenmaxxing,” pitching real-world AI value over hype
In comments highlighted by Yahoo Finance, Palantir’s CEO argues that investors should separate durable, use-case-driven AI progress from speculative “token industrial complex” narratives.
FTC and 22 states sue Amazon, alleging inflated prices in online ads scheme
The Federal Trade Commission and a coalition of states filed a lawsuit accusing Amazon of misleading advertising customers and defrauding them through inflated ad pricing. Amazon has not been found liable, and the company’s response was not included in the announcement referenced by the reporting.