The National Aeronautics and Space Administration (NASA) and The Boeing Company have established a revised flight schedule for the CST-100 Starliner spacecraft, targeting an uncrewed test flight in late 2026 and a crewed mission in 2028 while initiating a transition to a new launch vehicle.

Announced in a press release on September 28, 2026, the updated development plan outlines Boeing’s recovery path following the 2024 Crew Flight Test mishap. The strategy includes hardware modifications to the spacecraft, a re-designation of upcoming flights, and the certification of the United Launch Alliance (ULA) Vulcan Centaur rocket for future human spaceflight missions.

Revised flight schedule and hardware modifications

NASA and Boeing are targeting a launch window of December 2026 or January 2027 for the uncrewed Starliner-1 mission to the International Space Station (ISS). This flight will serve to validate recent thermal modifications and gather performance data before astronauts are placed back on board.

Historically, Starliner-1 was intended to be the first operational crewed mission following the initial test flights. The re-designation reflects the uncrewed nature of the next flight following the 2024 anomalies.

NASA Administrator Jared Isaacman stated that the agency is starting with an uncrewed mission to validate spacecraft improvements and gather necessary flight data.

“From there, we will use what we learn, continue implementing the corrective actions identified by our Program Investigation Team, and complete the testing and certification required for crewed flight,” Isaacman said.

Following the uncrewed test, the agencies are targeting 2028 for the crewed Starliner-2 mission. NASA astronaut Warren “Woody” Hoburg has been assigned as the commander for this flight.

Addressing the 2024 Crew Flight Test anomalies

The revised schedule follows the 2024 Crew Flight Test, which was classified as a Type A mishap. During that mission, the spacecraft experienced significant technical issues with its service module reaction control thrusters.

The official investigation concluded that the thrusters operated outside their engineering qualification due to a combination of thermal environment factors and design features, resulting in a loss of control during the flight. The spacecraft ultimately returned to Earth uncrewed. Astronauts Butch Wilmore and Suni Williams remained safely on the ISS, though their stay was extended by nine months due to the propulsion failures.

In February 2026, the NASA Program Investigation Team released 61 recommendations to address the technical issues. To meet these requirements, Boeing has implemented thermal modifications to the service module and an additional thruster valve design modification to address poppet seal extrusion. The spacecraft will also receive new crew module thrusters, updated batteries, and minor modifications to the parachute system.

Transitioning to the Vulcan Centaur launch vehicle

A critical component of the updated Starliner program is the transition to a new launch vehicle. The spacecraft currently relies on the ULA Atlas V rocket, which is out of production. ULA, a joint venture between Lockheed Martin and Boeing, has only six Atlas V rockets remaining in its inventory. All six are allocated to Boeing for Starliner missions.

To ensure the spacecraft has a launch vehicle for missions beyond the initial contract, NASA, Boeing, and ULA will work to certify the new Vulcan Centaur rocket for human spaceflight. The Vulcan Centaur is currently undergoing its own certification processes to replace the Atlas V.

Maintaining redundancy in low Earth orbit

The recovery of the Starliner program remains a priority for NASA as it seeks to maintain two independent commercial crew transportation providers. Currently, Space Exploration Technologies Corp. (SpaceX) and its Crew Dragon spacecraft serve as the only operational U.S. vehicle capable of ferrying astronauts to the ISS.

To support this goal, NASA and Boeing have modified the Starliner contract to add resources for human spaceflight certification and restore the fifth and sixth Starliner missions, which were previously made options.

According to reporting by Spaceflight Now, Dana Weigel, NASA Manager of the Low Earth Orbit Program, emphasized the necessity of this redundancy. “It’s always been the Commercial Crew Program’s goal to have two crew transportation providers to ensure commercial access to low Earth orbit,” Weigel said.

Weigel also noted in the NASA release that the next flight is a critical step toward full system certification. She stated that the agency will test the propulsion system through targeted demonstration objectives and disciplined operational controls, prioritizing the safety of the space station crew and the public.

AirPro News analysis

The decision to insert an uncrewed flight before resuming crewed operations underscores the severity of the 2024 anomalies and the extensive engineering work required to satisfy the 61 recommendations from the Program Investigation Team. The re-designation of the flights, shifting Starliner-1 from an operational crewed mission to an uncrewed test, reflects a necessary reset of the program’s baseline.

Tying the Starliner program’s long-term viability to the human-rating certification of the Vulcan Centaur introduces parallel development risks. If Vulcan certification encounters delays, Boeing’s ability to fulfill its commercial crew obligations beyond the remaining six Atlas V rockets could be constrained. This scenario would leave NASA reliant on a single provider as the ISS approaches its planned 2030 retirement and the agency looks toward future commercial space stations.

Firefly Aerospace and space data center startup Starcloud have signed a commercial agreement to deploy an artificial intelligence computing payload to lunar orbit as early as 2028. The mission will utilize Firefly’s Elytra orbital vehicle to host Starcloud’s SC-1L system, aiming to process massive volumes of data locally and transmit actionable insights back to Earth.

Announced in a September 30, 2026, press release, the collaboration seeks to validate the core capabilities required for a future lunar data center. By performing high-power computing in deep space, the companies intend to mitigate the severe bandwidth constraints that currently limit lunar data downlinks.

Validating lunar computing infrastructure

The integration of Starcloud’s SC-1L payload onto the Elytra vehicle represents a shift in how space missions handle data. Traditionally, spacecraft transmit raw data back to Earth for processing, a method constrained by limited deep-space network bandwidth. By pairing the SC-1L computing payload with Firefly’s Elytra vehicle and its Solux vision system, the mission will demonstrate how data can be captured, processed, and delivered directly from lunar orbit.

According to Firefly Aerospace, the Elytra vehicle is designed to remain in lunar orbit for five years to enable customer payload and imaging operations.

“We’re proud to collaborate with innovative customers like Starcloud and collectively take another step toward establishing the infrastructure that will power a permanent human and robotic presence at the Moon,” said Ray Allensworth, Vice President of Spacecraft at Firefly Aerospace.

Starcloud Cofounder and Chief Technology Officer Ezra Feilden noted that the company successfully validated its ability to run and train artificial intelligence models on enterprise-grade graphics processing units in low Earth orbit (LEO) before targeting the Moon.

“Space is the future of data centers, and the Moon is the next frontier for that vision,” Feilden said in the release. He added that the Elytra mission will demonstrate high-power computing and prove the ability to process massive volumes of data right where it is generated.

The push for space-based data centers

The agreement highlights a growing commercial sector focused on moving data infrastructure off-planet. Founded in January 2024 and headquartered in Redmond, Washington, Starcloud designs and deploys data centers in space to leverage the vacuum environment for cooling and continuous solar energy for power. This approach is designed to bypass the massive terrestrial energy and land constraints currently facing the artificial intelligence industry.

Starcloud has rapidly accumulated capital to fund this architecture. According to reporting by GeekWire, the company raised a $250 million Series A extension in August 2026, bringing its post-money valuation to $2.3 billion. The funding round included participation from NVIDIA, Cisco Investments, and Benchmark.

The company previously demonstrated its hardware in November 2025 by launching Starcloud-1, a satellite equipped with an NVIDIA H100 GPU, to validate artificial intelligence computing in LEO.

Firefly Aerospace has also been integrating advanced computing into its platforms. In April 2026, the Cedar Park, Texas-based manufacturer announced a collaboration with NVIDIA to embed the Jetson edge artificial intelligence platform on its Elytra spacecraft. That system is designed to process data for Firefly’s Ocula lunar imaging service, reducing the need to downlink raw image files.

Firefly’s expanding lunar campaign

The Starcloud payload will fly on Firefly’s third lunar mission, which is targeted for launch no earlier than 2028. The mission will also carry the company’s Blue Ghost lunar lander to the Moon’s Gruithuisen Domes under the National Aeronautics and Space Administration (NASA) Commercial Lunar Payload Services (CLPS) initiative.

Firefly, a publicly traded company (Nasdaq: FLY), has steadily built its lunar flight heritage. The company successfully landed its Blue Ghost Mission 1 on the Moon in March 2025. Its subsequent flight, Blue Ghost Mission 2, is targeted for no earlier than 2027 and will deploy the first Elytra vehicle to lunar orbit.

The Elytra vehicle serves as a transfer stage, communications relay, and payload host. It is equipped with the Solux vision system, formerly known as Sol3, which enables autonomous navigation and landing in environments without global navigation satellite system coverage.

AirPro News analysis

The agreement between Firefly Aerospace and Starcloud illustrates a critical transition in lunar exploration architecture. As government and commercial entities plan permanent lunar outposts, the communications bottleneck between the Moon and Earth has emerged as a primary operational constraint. By moving the computing power to the data source, operators can transmit only the processed outputs, drastically reducing bandwidth requirements.

Furthermore, Starcloud’s $2.3 billion valuation and its backing by major terrestrial hardware providers like NVIDIA suggest that space-based data centers are no longer viewed purely as aerospace research projects. We are seeing the terrestrial cloud computing industry recognize orbital infrastructure as a viable solution to Earth-bound power and thermal limitations. If the 2028 Elytra mission successfully demonstrates enterprise-grade computing in the high-radiation environment of lunar orbit, it could catalyze a new market for commercial deep-space data hosting.

This article summarizes reporting by Reuters by Joey Roulette, with additional information from SpaceX, Forbes, and Space.com.

Space Exploration Technologies Corp. (SpaceX) successfully launched its Starship vehicle on its 14th test flight on September 28, 2026, marking the heavy-lift rocket’s first successful insertion into Earth orbit and its first operational payload deployment.

Lifting off from the company’s Starbase facility near Brownsville, Texas, at 12:46 UTC (7:46 a.m. local time), the 40-story vehicle carried 26 Starlink V3 satellites. According to official statements from SpaceX, reaching orbit represents a critical transition for the program from passively safe suborbital developmental testing to operational spaceflight.

Orbital profile and payload deployment

The flight plan called for Starship to reach an altitude of 275 kilometers and complete six Earth orbits. Following a 10-hour orbital voyage, the upper stage was scheduled to splash down in the Pacific Ocean west of Chile.

During the ascent phase, the Super Heavy booster experienced a premature shutdown of a single Raptor engine, according to reporting by Forbes. The vehicle’s flight computer compensated for the loss of thrust, allowing Starship to successfully reach its target orbit without compromising the primary mission objectives.

The mission also served as the inaugural deployment of the company’s next-generation communications satellites. SpaceX noted that deploying the V3 satellites will deliver a payload designed to “dramatically expand connectivity speeds and reliability around the world.”

Regulatory approval and Artemis program implications

Prior to Flight 14, the Federal Aviation Administration (FAA) issued modified launch licensing to permit the orbital attempt. SpaceX had previously restricted Starship to suborbital trajectories to maximize public safety while gathering flight data.

The successful orbital insertion serves as a prerequisite for future lunar missions. The National Aeronautics and Space Administration (NASA) has contracted Starship as the initial crewed lander for the Artemis program, which targets establishing a base near the lunar south pole.

In a press release, SpaceX stated that achieving orbit allows the next phase of developing the vehicle “to be fully and rapidly reusable” to begin.

AirPro News analysis

We view the transition from suborbital testing to orbital payload delivery as a fundamental shift in Starship’s commercial viability. The successful deployment of Starlink V3 satellites demonstrates that the vehicle can now generate internal revenue and build out SpaceX’s proprietary infrastructure while continuing its development toward human spaceflight. The engine anomaly, while minor in the context of a successful orbital insertion, highlights the ongoing reliability challenges inherent in the 33-engine Super Heavy booster design. Consistent engine performance will be a primary focus for regulators before crewed Artemis missions can proceed.

Sources: SpaceX