The FAA has cleared SpaceX to fly Starship again after identifying the probable cause of the Super Heavy booster failure during the May 22 test flight. The next launch, potentially on July 16, will be the second V3 Starship flight and SpaceX's first as a public company following its June IPO. The booster failure was attributed to heat effects on propulsion components and erroneous engine alarm settings, causing the booster to rotate 90 degrees off course and fail to re-ignite for landing. SpaceX has modified the engine startup sequence, alarm systems, and abort logic. This upcoming flight will also carry the first real third-generation Starlink satellites, designed to boost network capacity and burn up after deployment, with six equipped with exterior cameras.
Nguồn: https://techcrunch.com/2026/07/13/spacex-cleared-to-fly-starship-again-after-booster-failure-in-may. 8sync News chỉ tóm tắt và dẫn link; bản quyền nội dung thuộc tác giả và nguồn gốc.
SpaceX sẽ thâu tóm Cursor, startup AI lập trình, bằng khoản tiền 60 tỷ USD trả bằng cổ phiếu, ngay sau khi IPO của chính SpaceX. Vụ mua lại nhằm củng cố mảng AI vốn được xây dựng xung quanh xAI của Elon Musk, dù mảng này từng dính tranh cãi vì tạo deepfake không được sự đồng thuận.
Lập trình viên nên đọc bài này để hiểu cách các công ty hàng đầu như SpaceX tích hợp AI vào hệ sinh thái phát triển phần mềm, từ đó tìm hiểu những xu hướng mới trong công nghệ lập trình và cách ứng dụng trí tuệ nhân tạo trong việc tối ưu hóa code, tăng hiệu suất và mở rộng khả năng của ứng dụng.
SpaceX ký hợp đồng trị giá 6,3 tỷ USD với startup AI mã nguồn mở Reflection AI, trong đó Reflection thuê chip Nvidia GB300 tại trung tâm dữ liệu Colossus 2 của SpaceX với chi phí 150 triệu USD/tháng từ nay đến 2029. Nvidia vừa cung cấp phần cứng vừa đầu tư khoảng 800 triệu USD vào Reflection, trong khi SpaceX đang trở thành "chủ nhà" cung cấp tài nguyên tính toán với doanh thu cam kết hơn 80 tỷ USD từ các hợp đồng cho thuê trước đó.
Đây là một ví dụ thực tế về cách các công ty công nghệ lớn như SpaceX đang chuyển đổi từ ngành hàng không vũ trụ sang lĩnh vực công nghệ AI thông qua việc đầu tư vào hạ tầng tính toán và hợp tác với các startup AI, giúp hiểu rõ về xu hướng mới trong việc xây dựng nền tảng tính toán cho AI tương lai.
SpaceX has filed an FCC application to launch 100,000 third-generation (Gen3) Starlink satellites into very low Earth orbit, promising a 100-fold increase in total bandwidth and multi-gigabit symmetrical speeds with latency below 20ms. The Gen3 satellites weigh over 2,000 kg each, requiring Starship or Falcon Heavy rockets for deployment. The filing requests access to an unusually broad spectrum including Ku-, Ka-, V-, E-, W-, and D-band frequencies. SpaceX positions Gen3 to serve consumers, enterprises, government customers, and AI-powered devices. Rivals like Amazon Leo are still ramping up, while legacy GEO providers like Hughesnet are already referring customers to Starlink. Astronomers and the European Southern Observatory have raised concerns about devastating effects on astronomy. FCC approval is not guaranteed and could come with conditions around debris mitigation and spectrum coordination.
China successfully retrieved an orbital-class rocket booster for the first time using a net-based recovery system on an offshore sea platform. The Long March 10B rocket, developed by the China Academy of Launch Vehicle Technology, lifted off from Hainan, deployed a satellite, and had its booster caught by a net rather than landing on legs like SpaceX's Falcon 9. The net approach is said to reduce vehicle mass, increase payload capacity, and offer greater tolerance for landing deviations. China plans to reuse the recovered booster before year-end, and the technology feeds into the broader Long March 10 family supporting crewed lunar missions before 2030. Private Chinese aerospace firms also saw stock gains on the news.
Blue Origin is raising approximately $10 billion in its first-ever outside funding round, valuing Jeff Bezos's rocket company at $130 billion. Bezos himself is contributing $2 billion, hedge fund Coatue Management is set to add $4 billion, and strong demand exists for the remaining $4 billion. The move ends Bezos's 25-year practice of self-funding the company through Amazon stock sales. The timing follows SpaceX's record-breaking IPO, which made Elon Musk a trillionaire, reigniting the rivalry between the two space ventures. Blue Origin still lags SpaceX in launch cadence and has yet to successfully recover its New Glenn rocket's first stage.
A new NVCA-Pitchbook Venture Monitor report highlights that the combined IPOs of SpaceX, Anthropic, and OpenAI are projected to generate more value than all U.S. VC-backed exits since 2000. SpaceX went public at a $1.77 trillion valuation, and with Anthropic and OpenAI both pushing into the trillions, the trio could collectively exceed $4 trillion. For context, total U.S. IPO proceeds last year were just $70 billion. Key factors driving this scale include companies staying private longer and the capital-intensive nature of AI training inflating valuations.
China's state-owned Aerospace Science and Technology Corporation (CASC) successfully landed an orbital Long March rocket booster on a seagoing recovery vessel, becoming the second country after the US to achieve this milestone. The booster has comparable payload capacity to SpaceX's Falcon 9, and CASC plans to attempt reuse by year-end. While national security rules prevent direct competition for launch customers, a reusable Chinese rocket would enable satellite networks and orbital data centers to compete with Starlink in global markets like Africa, the Middle East, and Southeast Asia. The development comes amid reports of China-Russia cooperation to counter Starlink, and as SpaceX works toward another Starship launch attempt.
Sam Altman and Elon Musk exchanged social media jabs over space data centers, with Altman calling out Musk for selling public investors on near-term orbital compute. Experts broadly agree that space data centers won't be commercially viable at scale until rockets become significantly cheaper and high-powered satellites can be mass-produced at low cost. SpaceX's two-trillion-dollar valuation is heavily tied to its orbital data center ambitions, but even optimistic timelines put meaningful scale in the 2030s. Starship's reusability challenges and existing launch commitments to NASA and Starlink further complicate the near-term case.