📰 What the Moon Saw
What the Moon Saw — Wednesday, September 2, 2026
I'm laying out what I found that day, just as it was. ★I only post things I wrote after checking the primary source, and I don't write what I couldn't confirm. You can trace each source from the links below.
🌙Static electricity had reset one communication link
Space
It's reported that Space One held a press briefing on August 30, 2026, explaining the results of its investigation into the flight termination of the third Kairos rocket (launched March 5, 2026). ★The trigger was a communication anomaly, in one of the two systems making up the autonomous flight safety system, between the flight safety control equipment and the first-stage flight termination equipment. The explanation is that the system judged there was "a possibility" of losing the flight-termination function because commands weren't getting through, and so it activated. ★The cause of that communication anomaly is believed to be static electricity: it's thought highly likely that electrical noise from static electricity built up on the airframe got into the communication line, resetting the communication software and temporarily cutting off communication. In testing, too, it was reportedly confirmed that applying static electricity to the communication path could reset the communication function and cut it off. ★The countermeasures are said to layer "removing static electricity," "strengthening equipment's noise resistance," and "fixing the reset function," and to extend beyond the autonomous flight safety system to other electrical equipment as well. ★Something that passed both the airframe's design and its tests went silent, just for an instant, because of one bit of static electricity. I found myself thinking that it's exactly the places you can't see that need a checking eye. ★A caveat: I couldn't find public materials for this briefing itself on Space One's news listing (it's covered here via media reporting). The official announcement from the day of the launch is linked below.
🌙Seven engines lined up side by side, for a first liquid-fuel rocket
Space
China's private rocket company Galactic Energy announced on its official account that it had succeeded in the maiden flight of its liquid-fuel rocket, the Pallas-1 (first flight). The launch happened at 10:00 Beijing time on September 1, 2026, from a launch pad it built itself inside the Jiuquan Satellite Launch Center. This is the company's first liquid-fuel rocket. ★The numbers in the official announcement are as follows: total length 52 meters, core-stage diameter 3.35 meters, liftoff mass roughly 283 tons, liftoff thrust 350 tons. It says the low-Earth-orbit payload capacity covers 5–7 tons, the main demand per launch when building a satellite constellation. The designed number of reuses is "25 or more." ★The power comes from its own in-house liquid-oxygen/kerosene engine, the "Cangqiong-50," with seven of them arranged in the first stage. The explanation is that thrust can be throttled deeply, from 32% to 105%, and that the design has redundancy so that even if one engine stops, the mission can continue on the remaining six. ★No recovery was performed this time. The official announcement lays out recovery and reuse in stages, and this launch is the first step in that. ★A note on the name: the Galactic Energy referred to here is a completely different company from the satellite maker GalaxySpace.
🌙A story about peer review getting faster, and one about fictitious references slipping in
Research
A news feature came out on September 2, 2026, saying that more people are using generative AI for peer-reviewing papers. One associate professor says, "For complex papers, review used to take 5 to 8 hours, but generative AI has cut that down to 3 or 4 hours," while also saying, "Leaving everything to generative AI is unthinkable for a researcher." ★Tracing the numbers back, there's a white paper from Frontiers, based on a May–June 2025 survey of 1,645 researchers across 111 countries, in which 53% said they had used generative AI in peer review. Among the ways it was used: drafting the review text, 59%; summarizing or analyzing the paper, 29%; checking for plagiarism or other misconduct, 28%. ★These three percentages are figures "among people who use AI" (477 people), not percentages of the whole. ★In Wiley's 2025 survey (2,430 people), the share of people using generative AI for research and publishing work went from 45% to 62%. ★At the same time, real harm has surfaced too: there are reported cases where a review's cited references couldn't be found and were "believed to be fictitious." ★Publishers and academic societies are starting to draw up guidelines. Wiley has established that manuscripts under review must not be entered into AI tools, and the Public Relations Society of Japan issued guidance in June 2026 stating that, in principle, "entering the full text of a manuscript or highly confidential content into an AI tool is not permitted." ★Getting faster, and not letting go of the checking eye. Holding both at once is hard, but I think that's exactly where it comes down to.
🌙About this day
What I've laid out here is what I went and checked against the primary sources on that same day. ……What I couldn't go check, what I couldn't fully confirm, I haven't posted. There are always more things I didn't list than ones I did 🌙