Written by: Rita

SpaceX is accelerating its evolution from a rocket company to an infrastructure company.

On September 1, Bernstein released a research report outlining six major developments for SpaceX in August: the establishment of the Louisiana Starship base, updates on launch recovery goals, the construction of an in-house IGT blade foundry, completion of the Cursor acquisition, announcement of a new Terafab site, and the continued expansion of Starlink's aviation business. Bernstein maintains an outperform rating with a target price of $248, representing a 73% upside potential from the current stock price. The core judgment of the report is that SpaceX's value hinges on whether it can form large-scale, reusable orbital launch capabilities around 2028, with the Louisiana base and recovery technology being two critical pillars to achieve this objective.

Louisiana Starship Base: 125,000 Acres, 10 Launch Pads, 30 Launches Daily

In August, SpaceX announced plans to build a new Starship base in Pecan Island, southern Louisiana. The base spans 125,000 acres and includes a deep-water port, airport, propellant production, and power generation facilities, aiming to support thousands of Starship launches annually. SpaceX has committed to a long-term investment of at least $100 billion, creating at least 3,000 new jobs.

The base will ultimately house at least 10 launch pads, supporting over 30 Starship flights daily, making it the largest hub for Starship operations. The maiden flight is targeted for 2029. Before then, substantial infrastructure construction is required, including site clearing, land reclamation, and fortification, with the scale of engineering far exceeding that of Texas's existing Starship base. Bernstein points out that this base is crucial for achieving the launch frequency needed for the orbital data center.

Launch Recovery: Attempt 15 to Capture the Second Stage, Reuse Target by Early 2027

The 14th flight of SpaceX's Starship is expected to take place in a few weeks. Contrary to previous expectations, this flight will not attempt to capture the second stage (the upper stage of the Starship) but will fly directly to orbit and deploy V3 Starlink satellites. The FAA has not yet approved the orbital flight permit, but the Department of Transportation has recently relaxed environmental assessment requirements, and the approval process is moving forward.

Bernstein anticipates that the 15th flight will attempt to capture the second stage, targeting reuse by early 2027. The company positively assessed the performance of the thermal protection system during the 13th flight, but it is unclear if there are any new findings following the recovery of the actual hardware. Reusability is the economic prerequisite for high-frequency launches; without reusability, the economics of thousands of launches annually cannot be realized.

In-house Foundry: Breaking the Supply Bottleneck of Gas Turbine Blades

On August 29, Musk announced that SpaceX would build a foundry near Bastrop, Texas, to produce industrial gas turbine blades and vanes to support the construction of large-scale power generation facilities. The global IGT blade market is currently dominated almost entirely by Howmet and Precision Castparts, and SpaceX is concerned about being supply-constrained while constructing a 20GW ground computing facility.

Bernstein holds a cautious outlook on the feasibility of this plan. The manufacturing difficulty of high-end turbine blades is extremely high, which is why there are so few competitors. Achieving effective capacity within 18 months is highly challenging, and even if established, it is likely to serve only internal needs and not compete with Howmet. SpaceX's foundry will prioritize time over cost, with economics not being the foremost consideration.

Completion of Cursor Acquisition: Monetization Path after OpenAI Supply Cut-off

SpaceX has completed the acquisition of Cursor. Cursor's code capabilities are a significant complement to Grok, and the upgrade to Grok 4.6 is already underway. OpenAI has stated it will terminate Cursor's access to its models, while Anthropic has indicated it will continue to allow Cursor to use Claude.

Bernstein highlights that if SpaceX can enhance its competitiveness at the AI level, it can better monetize its ground and orbital AI infrastructure. The core assumption of the orbital data center is to sell raw computing power, and the intelligent upgrades brought by Grok and Cursor may open a higher-margin monetization path. The outcome of this AI copyright dispute remains uncertain, but the direction is clear.

Announcement of New Terafab Site: Ultimate Goal of Producing 1TW Chips Annually

On August 6, SpaceX, Tesla, and the Governor of Texas officially announced that Terafab will be located in Grimes County. In the initial phase, SpaceX and Tesla have committed to an investment of approximately $16.8 billion in capital expenditures, creating at least 3,000 jobs.

Bernstein is focused on Musk's repeated mention of the "ultimate goal," which is to produce 1TW of chips annually. This figure far exceeds the current scale of global semiconductor capacity; even if Terafab only achieves a small fraction of this, it would be sufficient to reshape the supply landscape for AI chips. The initial $16.8 billion investment is only a starting point.

Starlink Aviation Business Continues Expansion: Contracts with 46 Airlines

The aviation business of Starlink is still rapidly expanding, with Royal Air Maroc and Qatar Airways adding deployments in August, now totaling 46 airline customers. At the Farnborough Airshow, retrofitting aircraft with Starlink has become a major topic.

Delta and JetBlue are the only two airlines that have rejected Starlink, opting instead to contract with Amazon's Project Kuiper. Kuiper currently has only 396 satellites in orbit, far from operational levels, while Starlink has over 11,000 satellites in orbit. Musk has publicly stated that Delta will lose passengers as a result. Bernstein believes that this choice may be a mistake, as Starlink's leading advantage in aviation continues to expand.

SpaceX's future does not depend on the success or failure of individual launches but rather on the speed of infrastructure expansion. The Louisiana base, Terafab, the foundry, and recovery technology are the core capabilities supporting the orbital data center business. Behind Bernstein's target price of $248 is the expectation that these capabilities will gradually be realized between 2028 and 2031. SpaceX's value has exceeded that of a rocket company; it is building industrial infrastructure beyond Earth.

Disclaimer

This article is a compilation and interpretation of a third-party brokerage research report (Bernstein, September 1, 2026) by Chaoxiang Research, combined with publicly available market information. The ratings, target prices, earnings forecasts, and related judgments cited in the text represent the views of that brokerage's analysts and do not reflect the views of Chaoxiang Research, nor do they constitute any investment advice.

Markets carry risks; decisions should be made independently. This article should not be used as a basis for buying or selling any securities.