What the FCC
Actually Approved
Almost every report said the FCC approved the first commercial space mirror. It didn’t. It approved a radio licence — and then said in plain terms that the mirror is none of its business. Here is what the order says, read line by line.
The scope of the grant
Strip away the coverage and read the grant itself. Under Scope of Grant, order DA-26-706 authorises exactly one thing:
Authority to deploy and operate a space station using radio frequencies in the S-, X-, and UHF- bands.
That is a radio licence. It permits Earendil-1 to talk to the ground: telemetry, tracking and command on UHF and S-band, and payload data downlink on X-band. The frequencies are itemised to the kilohertz. The reflector is not authorised, because — in the Commission’s own view — it was never the Commission’s to authorise.
The jurisdiction gap
Most coverage missed this — though not all of it. Engadget reported it accurately, noting that the satellite needed FCC approval because of the radio spectrum it uses, and quoting the Commission’s finding that the health and environmental concerns are unrelated to its role in authorising spectrum. The distinction matters, so it is worth stating plainly: the Space Bureau did not weigh the mirror’s risks and find them acceptable. It found that the mirror sits outside what it is allowed to consider at all.
And more directly still: the Bureau states that it does not have regulatory authority over the licensing and operation of a solar reflector, that impacts of the reflector are beyond the scope of the Commission’s authority to regulate, and that consequently the reflector’s operation would not be reviewed as part of the Bureau’s public interest analysis.
The legal engine driving this is worth naming: the order cites Loper Bright Enterprises v. Raimondo (2024) — the Supreme Court decision that ended Chevron deference. In the post-Chevron environment, agencies read their own authority narrowly rather than expansively. The Bureau adds that imposing novel, service-wide obligations through a single licence condition would be procedurally inappropriate; that kind of rule, it says, belongs in a notice-and-comment rulemaking.
So the finding is not “the mirror is safe.” The finding is “the mirror is not ours to judge.”
Then who does regulate the beam?
On the record as it stands: no one has claimed it. The FAA reviews the launch. The FCC licenses the radio and reviews orbital debris. The FCC has now said the optical behaviour of the reflector falls outside its remit. No other US agency has asserted jurisdiction over deliberately directing sunlight at the ground at night.
That is the actual story of 9 July 2026. Not that a regulator approved a space mirror — but that a novel technology reached orbit-readiness through a gap where no regulator claims authority over its principal effect.
Why the AAS petition was denied
The American Astronomical Society, representing more than 8,500 members, filed the sole formal petition to deny. It was DENIED. But read why: the Bureau rejected the astronomy arguments primarily because they concerned the reflector, which it had already placed outside its authority — not because it examined and dismissed the science.
The Bureau did add an alternative finding: even if it did have authority, it would still grant, on the basis that Earendil-1 is a single, short-duration test rather than the constellation most objections addressed. It repeatedly sets aside constellation-scale concerns as speculative, noting that any future application would be reviewed on its own merits.
Two further filings — from the Asociación Argentina de Astronomía and the Sociedad Española de Astronomía — were titled as petitions to deny but lacked the required affidavits and certifications of service, so the Bureau treated them as informal objections instead. A detail worth knowing: procedure filtered the opposition as much as substance did.
What the IAU asked for — and didn’t get
The International Astronomical Union requested five specific licence conditions. All were declined as redundant with Reflect Orbital’s voluntary commitments. They asked for scattered-light profiles across the visible spectrum; a viable plan for controlling beam pointing in all mission phases; no pointings within 160 km of observatories on the Minor Planet Center’s register; continued consultation with the NSF and AAS; and — the one that matters most to anyone trying to watch this thing —
The IAU wanted that schedule published to prevent eye injury among amateur astronomers. The FCC declined to require it. So there is no regulatory obligation on Reflect Orbital to tell the public where the beam is going or when. Whether an illumination schedule ever becomes public rests entirely on the company choosing to publish it.
What actually does bind Reflect Orbital
The conditions that exist are real, but they are almost all about radio and debris, not light:
- Transmissions are limited to windows when Earendil-1 is visible to named, coordinated earth stations — a list that includes Mingenew, Western Australia and Awarua, New Zealand, alongside Svalbard, Troll in Antarctica, Punta Arenas and others.
- Detailed duty-cycle caps and conjunction-angle cut-outs protect specific NASA and NOAA missions — transmissions must cease near EZIE, TRACERS, PUNCH, TIMED and JASON-3 geometries.
- Orbital debris: the satellite carries propulsion for collision avoidance, must deorbit within a year of mission end, and its calculated human casualty risk is 1 in 119,400 — only the titanium propellant tank is expected to survive re-entry.
- Reflect Orbital must operate in accordance with the commitments it made on the record. That is the hook that makes its promised NSF and NASA coordination to protect optical astronomy binding — not a condition the FCC imposed, but a promise the licence incorporates.
- Initial orbital parameters must be reported to the FCC, AFSMO, NASA and NOAA within 30 days of launch.
The dates that actually matter
The order sets a schedule almost nobody has reported. Two of these fall within weeks.
- 8 AUG 2026Deadline for Reflect Orbital to decline the grant as conditioned. Thirty days from release. Silence constitutes formal acceptance.
- 10 AUG 2026Surety bond due under 47 CFR § 25.165(a). Miss it and the authorisation becomes null and void. This is the nearest hard gate on the licence.
- 9 JULY 2032The actual launch deadline in the licence. Six years. The FCC applies no near-term pressure whatsoever — “later in 2026” is Reflect Orbital’s ambition, not a regulatory requirement.
- LAUNCH + 2 YRLicence term. Roughly one year of operations plus deorbit. Earendil-1 is authorised as a short-duration demonstration, and then it is over.
That 2032 date deserves emphasis. Coverage has treated the approval as the last gate before an imminent launch. In licence terms it is nothing of the sort: Reflect Orbital could take six years and remain fully compliant. The pressure to fly soon is commercial and competitive, not regulatory.
The grant specifies an operating orbit of 625 km at an inclination of 88° ±2°. That is not an 88° near-polar orbit. Sun-synchrony at 625 km requires roughly 97.9°; an 88° orbit precesses about −0.25° per day, the wrong way and by the wrong amount, and its local solar time would drift through a full cycle in under a year.
A figure of 98° ±2° would bracket 97.9° precisely and precess at +1.0° per day — almost exactly the +0.9856° sun-synchrony needs. The stated tolerance only makes sense around 98. We read this as a transcription error in the grant rather than a real deviation, and we have not changed our orbital figures on that basis. We note it because the licence is the primary source, and where the primary source appears wrong, that is worth saying out loud rather than quietly picking whichever number suits.
What Reflect Orbital said
Ben Nowack’s statement on the grant runs to three sentences. He thanked the FCC for recognising the importance of testing novel technologies in space, said the company was excited to demonstrate the technology, and — the phrase worth noting — called the licence “the first step toward rigorously testing our technology’s efficacy and the safeguards we have developed.”
The first step. Reflect Orbital’s own framing is closer to the licence than the headlines were. The company has not publicly addressed the jurisdictional finding at all. When Engadget asked it to respond to the American Astronomical Society’s criticism, Reflect Orbital did not make its own argument — it pointed the reporter to the FCC’s opinion, and specifically to the Commission’s reasoning that eye injury would require someone to stare at Earendil-1 through a telescope larger than 12 inches at precisely the wrong moment. It is letting the regulator’s dismissal do the arguing.
The bottom line
Reflect Orbital holds a valid radio licence for a single demonstration satellite, granted on the finding that authorising spectrum for emergent space activity serves the public interest — the Commission called the technology “potentially groundbreaking.” The astronomical community’s objections were not weighed and rejected on their merits. They were ruled out of scope.
Every substantive question the objectors raised — skyglow, eye safety, wildlife, aviation, the constellation to come — remains open. What changed on 9 July is that a US agency stated clearly that answering them is not its job.
SOURCE: FCC Space Bureau, Reflect Orbital Inc., Order and Authorization and Memorandum Opinion and Order, DA-26-706, ICFS File No. SAT-LOA-20250701-00129, Call Sign S00711, adopted and released 9 July 2026. Read the order in full at docs.fcc.gov. Quotations are from the order; emphasis ours. Analysis is our own.
See also: how FCC satellite licensing works, the scientific opposition, and the launch-date analysis.