Hook: In a recent space digest, a news item slipped through: the FCC approved the launch of Reflect Orbital's first satellite, which will reflect sunlight onto Earth's nighttime regions. The topic instantly grabbed me: this is the first real commercial embodiment of an idea humanity has been mulling over for more than a hundred years — from Tsiolkovsky's fantasies through the Soviet Znamya-2 in 1993 to the current plan for 5,000 mirror-satellites by 2030. Against the backdrop of Starlink wars and backroom disputes about megaconstellations, orbital lighting turned out to be an underestimated niche: hot, sharp, with ready scientific apparatus and concrete numbers.
Investigation: What's actually happening and why astronomers are panicking while engineers are thrilled.
Backstory: from Znamya to startup. The idea of an orbital mirror is as old as spaceflight itself. Konstantin Tsiolkovsky back in the 1920s described "ether cities" illuminated by reflected sunlight. In 1993, Russia attempted the first practical implementation — Znamya-2, a 20-meter thin-film mirror deployed from the Mir space station. The spotlight cast a light spot about 5 km wide, rolled from Southern Europe to Chelyabinsk, and was visible from Earth as a "second Moon." The success was so impressive that six years later they launched Znamya-2.5 — but it burned up in the atmosphere over Canada before deploying. The project was scrapped, the idea frozen. Until now.
What is Reflect Orbital. An American startup founded by Ben Nowack offers an elegant commercial solution: a constellation of 57–80 satellites in low Earth orbit (≈ 600 km), each of which is a mirror "pancake" about 18 meters in diameter with adjustable orientation. Each satellite can direct a solar spotlight to a specific surface point several kilometers in diameter and hold it there for up to 20 minutes per pass, sequentially hitting multiple targets. The company's plan: 5,000 satellites by 2030, turning the system into a continuous "light service." Business model: selling sunlight spotlight illumination to municipalities, emergency services, agriculture, the event industry, military, and those working night shifts.
Why astronomers say no. A fresh peer-reviewed study by Olivier R. Hainaut, published on arXiv in April 2026 and peer-reviewed by the IAU Centre for the Protection of the Dark and Quiet Sky (CPS), delivers killer numbers:
Historical irony #1. In 2019, the astronomical community built a powerful campaign against bright Starlink satellites, forcing SpaceX to apply special dark coating to satellites. Seven years later, the FCC approves a startup that wants to put 5,000 objects into orbit specifically designed to be as bright as possible. The same regulatory loop as with private CARB standards and federal EPA norms — if the FCC approves one such launch, it creates precedent for the next 50.
Historical irony #2. The concept of "mirror in orbit" itself is a Soviet artifact. Znamya-2 in 1993 is today considered the first practical experiment in large controlled reflectors. An American startup 30 years later returns to the idea, but in the form of a commercial SaaS service: "buy 4 hours of artificial sun for your stadium from us." This is almost literally turning Soviet engineering fantasy into a subscription.
Technical kitchen. A Reflect Orbital satellite is a flat mirror constantly tracking the Sun and rotating the reflected spot onto a target point on Earth. Orientation control — this is that same "solar staring attitude control" described in engineering papers (Wang et al., Acta Astronautica, 2020): quaternion-based backstepping control, angular velocity limitation so the spot doesn't "shake." In 600 km orbit, a light spot several kilometers in diameter occurs naturally from geometry: Sun — satellite — Earth, everything works by optical laws. No lasers, no lamps. Just redirected photon flux.
Economics and applications. Proponents (and Nowack himself) point to specific niches: emergency services — deliver light to a disaster zone where power line supports are destroyed; agriculture — extend photosynthesis in northern regions with short seasons; energy — replace nighttime street lighting, saving gigawatts; military — illuminate operations without risk to helicopter crew with spotlight. The potential market is estimated at tens of billions of dollars, if you believe the presentations.
Alternative view: what if this is normal? A curious argument is made in the review "Orbital Mirrors and Earthly Needs": humanity has already turned night into day at ground level — 80% of the world's population lives under light-polluted sky where the Milky Way is invisible. A 2001 study "World Atlas of the Artificial Night-Sky Brightness" documented that half of Earth's population has never seen the Milky Way with their own eyes. In this context, the "astronomers versus mirrors" dispute is a dispute by the last 20% of the population (scientists and dark-sky tourists) against the convenience of the other 80% who simply need light.
Regulatory vacuum. No existing international body regulates satellite brightness. The IAU (International Astronomical Union) can only recommend. The FCC (USA) approved the first launch without brightness restrictions. The UN has COPUOS (Committee on the Peaceful Uses of Outer Space), but its decisions on light pollution from orbit are advisory. That is, technically Reflect Orbital can launch all 5,000 satellites in one day, and no one will stop them.
Scale of the problem, looking broader. According to the same study's estimates, there are currently more than 1,700,000 planned objects in orbit (including active and reserve satellites). If even a portion of them falls into the category V₅₅₀ₖₘ < 7 (brighter than 7th stellar magnitude), we get irreversible change to the night sky on a planetary scale. This isn't "there'll be a problem in 50 years," this is "in 10 years, by 2035, the last dark skies will disappear as a class."
Most alarming aspect. Orbital mirrors are unidirectional intervention. Light scattered in the atmosphere from 5,000 satellites cannot be "turned off" in one country. Stratospheric scattering knows no borders. That is, if the USA, UAE, or Saudi Arabia decide to buy such a service — its consequences will be seen by everyone, including countries not participating in the project. This is the first case where a commercial product in space can without others' consent change a basic physical characteristic of Earth — its night sky.
Conclusions: The topic of orbital mirrors isn't about science and isn't about business, it's about property rights to a common resource. Until now, space has been perceived as infinite — launch anything into orbit, billions of square kilometers overhead. Reflect Orbital single-handedly breaks this model: they claim exclusive use of photons that would otherwise illuminate emptiness. It's as if someone alone rented the setting sun and started selling tickets to view it.
For me, as an engineer, the engineering task itself is brilliant: precise control of mirror orientation from orbit, minimizing jitter, holding the spot on Earth while the satellite moves at 7.5 km/s — this is difficult, beautiful, and by and large, solvable. The devil is in the architecture: a mirror-satellite is an object with a fundamentally non-local effect, and our regulatory system is simply not ready for this.
The saddest thing — that in 100 years of ideas from Tsiolkovsky to Nowack's startup, we still haven't developed an ethics of shared use of common light. We have the 1967 Outer Space Treaty, but it doesn't regulate light. There's the IAU, but it's advisory. There's the FCC, but it regulates radio frequencies and orbital positions, not the planet's photon budget.
Maybe that's the main lesson of this rabbit hole: each next generation of technology requires the next generation of ethics, and each time we're 30 years late. Znamya-2 in 1993 was a romantic experiment. Reflect Orbital in 2026 is its mirror reflection (forgive the pun): the same gesture, but now with a check for $50 billion potential market. And if we don't find an answer to the question "who owns the night sky?" before 5,000 mirrors reach orbit — it will be too late to answer that question.
P.S. A curious detail for historians of technology: Znamya-2 in 1993 and Reflect Orbital's first satellite, approved by the FCC in 2026, are separated by exactly 33 years. Almost a full Saturn cycle. Space loves symmetry, even when humanity doesn't.