TL;DR
SpaceX's Starship reached orbit for the first time on its 14th test flight, deploying 26 Starlink V3 satellites and proving the rocket's commercial and strategic viability despite an engine failure that cut the mission from 10 hours to 3.
What happened
- September 28, 2026: Starship lifted off at 7:48 a.m. local time from Starbase, Texas, and achieved orbit at approximately 8:15 a.m.
- One vacuum engine failed mid-flight, briefly threatening the orbital attempt before engineers confirmed a corrective burn was still possible.
- 26 Starlink V3 satellites deployed at 269 km altitude, each from a rocket capable of carrying 60 V3 satellites per launch, roughly 20 times the capacity of a Falcon 9.
- Mission was cut to 3 hours instead of the planned 10, with splashdown in the Pacific Ocean rather than off Chile after six planned orbits.
- The Super Heavy booster completed a soft splashdown off the Starbase coast shortly after launch.
Why it matters
- Starlink V3 scale-up begins: One Starship carries 60 V3 satellites versus Falcon 9's 3, compressing the timeline to dominate global broadband. SpaceX already operates 11,000 active satellites versus Eutelsat OneWeb's 650.
- $2 trillion company, one critical rocket: Starlink is SpaceX's only profitable segment, and Starship is the sole path to the launch cadence needed to sustain and grow it.
- NASA Artemis IV dependency: SpaceX is preparing Starship for a lunar landing mission no earlier than 2028, competing directly with Jeff Bezos's Blue Moon for the contract.
- AI compute in orbit: COO Gwynne Shotwell confirmed plans to launch AI compute satellites aboard Starship in 2027, a direct signal that orbital supercompute is moving from concept to manifest.
- Reusability and reliability are not yet demonstrated: the booster splashed down rather than being caught, and the engine failure underscores that full certification remains ahead.
What to watch next
- Flight cadence: Musk previously targeted one Starship launch every two weeks by end of 2025. Watch whether 2026 closes with anything close to that rhythm.
- Booster catch and full reuse: The next flights must demonstrate the Super Heavy being caught and relaunched to validate the cost model that underpins everything else.
- AI compute satellite announcement: A 2027 launch of orbital inference hardware would mark a new competitive front. Watch for payload partnerships or SpaceX-internal compute contracts in the next 12 months.
Originally published on Present of AI, a daily source-linked AI news timeline. Read the full timeline or browse the open dataset.