Space Exploration Technologies Corp (NASDAQ:SPCX) next growth engine may depend as much on artificial intelligence as rockets, with Elon Musk betting that compute demand, satellite connectivity and in-house chipmaking can reinforce one another. Its third-quarter earnings call offered concrete targets for that ambition, but also left investors with a bigger question: How quickly can SpaceX turn its infrastructure spending into durable returns?

SpaceX's AI Spending and Returns

Musk's answers highlighted how aggressively SpaceX plans to expand its AI infrastructure, while CFO Bret Johnsen pointed to operational cash flow, debt markets and partnerships as potential funding sources.

1. Can SpaceX Turn AI Compute Into Sustainable Profits?

Musk said SpaceX aims to split its compute capacity roughly evenly between internal workloads and external customers. He also said contracts are renewing at "higher dollars per watt," arguing that improving AI models could increase the value of computing capacity.

His confidence was striking: "I don't think we're going to have a demand issue here." Investors still need clearer evidence of how that demand translates into sustained profits.

2. Can SpaceX Deliver 13 Gigawatts of Compute by 2027?

Musk said he was "highly confident" SpaceX could add 10 gigawatts of capacity next year, taking its total to roughly 13 gigawatts. His more ambitious target is considerably higher: 20 gigawatts of incremental power, which he called "difficult but not impossible."

SpaceX is manufacturing turbine components internally to tackle a bottleneck Musk said can delay deliveries by 12 to 18 months.

3. Can Terafab Begin Volume Chip Production by 2028?

Musk outlined a target for a functional test chip around May or June 2027, followed by potential volume production in late 2028. He acknowledged the dates were "insanely fast timeframes for chip production."

The milestone to watch is whether SpaceX can translate its chipmaking ambitions into working manufacturing capacity.

Starlink and Starship Milestones

4. Can Starlink Become a Hundreds-of-Billions-Dollar Business?

Musk said SpaceX sees "a clear path to several hundred billion dollars a year in connectivity revenue," but deferred a detailed explanation.

Investors still need to understand the assumptions behind that ambition and how spectrum, satellite capacity and wireless services will contribute.

5. When Will Starlink's V3 Satellites Improve Service?

Musk said users would need "a few hundred satellites" to feel the impact of V3, a milestone he expects to reach by mid-2027.

That timeline gives investors a measurable checkpoint for network improvements.

6. Can Starship Scale Launches From Louisiana by 2029?

Musk set an internal target for the first Louisiana launch in 2028 and said launches "at scale" in 2029 were highly likely.

Execution against those dates will help determine how quickly Starship infrastructure can support SpaceX's broader ambitions.

The Tesla Merger Question

7. Is a SpaceX-Tesla Merger Really Happening?

Musk said a merger "would be a good idea," citing potential cooperation in AI software, chip design, robotics and satellite connectivity. But he cautioned that treating a deal as certain would be "overstating it."

The independent committees have not agreed on terms, and Tesla Inc (NASDAQ:TSLA) needs approval from holders of 50% of all shares.

For investors, the next test is execution: compute capacity coming online, Terafab hitting manufacturing milestones, Starlink improving service and Starship meeting launch targets.

Musk has laid out an expansive vision, but progress against these measurable milestones will determine how much of that ambition becomes financial reality.