EN
Back to the archive

The archive · Space, Robots, Defence · Legal decision · 2016–2018

Swarm's rogue SpaceBee launch draws an FCC rebuke that imperils its satellite IoT bet

Stealth startup Swarm launched four tiny SpaceBee satellites after an FCC denial; the agency revoked its next mission, leaving the IoT network bet in limbo.

Swarm Technologies

The betThat satellite IoT can cost 400x less using tiny 0.25U SpaceBees, and that launching them despite an FCC denial would read as momentum, not defiance, once proven.Live

What the business is

Swarm Technologies builds small two-way communications satellites ('SpaceBees') that relay data from IoT sensors on ships, farms and remote sites to the internet, aiming to make satellite connectivity orders of magnitude cheaper than existing networks like Iridium.

Starting capitalOver US$220,000 in National Science Foundation grants as of March 2018; no disclosed venture funding while in stealth.

How it started

Founded in 2016 in Menlo Park by aerospace engineers Sara Spangelo (NASA JPL, then a Google X spacecraft lead) and Benjamin Longmier (who sold near-space balloon company Aether Industries to Apple in 2015), Swarm bet that pervasive IoT connectivity could be built from satellites one-quarter the size of a standard CubeSat. Its FCC application claimed the platforms were less than 1/10,000 the mass and power and 400 times cheaper than Iridium-class systems, and it planned to launch four 'BEE' test satellites plus two ground stations on India's PSLV.

What happened

The FCC dismissed Swarm's application in December 2017 on safety grounds: sub-1U objects are hard to track, so the agency could not find the launch in the public interest. Yet on 12 January 2018 four SpaceBee satellites flew on the PSLV anyway, in orbits matching Swarm's own filing, in what would be — if confirmed — the first ever unauthorized launch of commercial satellites. A follow-up application for four larger 1U satellites on a Rocket Lab mission won FCC approval in weeks, and Swarm filed for a market trial with up to 500 uplink gateways, two Fortune 100 paid pilots, 125+ potential end users, a NASA Ames partnership and a second NSF grant. Two days later the FCC revoked the Rocket Lab authorization and said it would weigh the apparent unauthorized launch in deciding Swarm's qualifications as a licensee.

How it ended up

As of 2018-03-09 the FCC had revoked Swarm's follow-up mission authorization and put its market trial in jeopardy, and Swarm had not commented publicly; whether it could keep its license — and its network — was unresolved.

Background

Swarm Technologies, a stealth Menlo Park startup founded in 2016, bet that a space-based Internet of Things network did not need Iridium-scale satellites. Its tiny 0.25U 'SpaceBee' satellites — about a quarter of a standard CubeSat — would carry data from solar-powered ground gateways that collect Bluetooth, LoRa or Wi-Fi signals from ships, tractors and sensors, and relay it to the internet over VHF. Swarm's own FCC filings claimed the platforms were 400 times cheaper than existing satellite systems like Iridium, at less than 1/10,000 the mass and power.

The company was the work of Sara Spangelo, a NASA Jet Propulsion Laboratory veteran who later led a spacecraft concept at Google X, and Benjamin Longmier, who sold his near-space balloon company Aether Industries to Apple in 2015. In April 2017 Swarm asked the FCC for permission to launch four BEE test satellites and two ground stations on India's PSLV rocket. The FCC dismissed the application that December: objects smaller than 1U are difficult to track, and the agency said it could not conclude the launch was in the public interest.

Swarm appears to have launched anyway: on 12 January 2018 four SpaceBees reached orbit on the PSLV, in orbits matching its own application — the first ever unauthorized launch of commercial satellites, if confirmed. Swarm then won fast FCC approval for four larger 1U satellites and filed for a US market trial with up to 500 uplink gateways, two Fortune 100 paid pilots, 125+ potential end users and a NASA Ames partnership. Two days later the FCC revoked the Rocket Lab authorization and opened a review of whether the unauthorized launch affected Swarm's qualifications as a licensee.

The case also exposed a gap in the new space economy: ride-share launchers and foreign providers may not reliably verify US licenses for secondary payloads, and even a tiny satellite can be catastrophic in a collision. As of 2018-03-09 Swarm was still in stealth, not responding to Spectrum, with its market trial and network future hanging on the FCC's review.

What has to be true

  • The IoT needs connectivity where there are no networks — at sea and in rural areas — and Iridium-class prices blocked the market, making a 400x cheaper satellite a plausible unlock.
  • Miniaturization was the wedge: 0.25U satellites flying four-for-the-price-of-one as secondary payloads made the network cheap to build and test.
  • Swarm treated the FCC denial as a formality and launched anyway — a Silicon Valley speed bet that regulators would not stop a working demo.
  • Real demand predated the network: two Fortune 100 paid pilots, 125+ potential end users, US military interest and NSF funding were all in place before the FCC's revocation.

What can be applied

A regulator's no is a constraint, not a suggestion: Swarm's unauthorized launch turned a fixable denial into an existential license fight, because speed cannot substitute for permission.

Aftermath

As of 2018-03-09 Swarm had not commented on Spectrum's findings. The FCC's Experimental Licensing Branch had revoked authorization for the four-satellite Rocket Lab mission and opened a review of whether the unauthorized launch affected Swarm's qualifications to be a licensee; the market-trial application — up to 500 US uplink gateways for two Fortune 100 pilots — hung in the balance. Spaceflight said it had never knowingly launched a customer denied an FCC license; neither it nor ISRO would confirm license checks. No later outcome is established in this source.

Sources

spotted an error? The archive wants to know.

Your turn

You just read one. Describe what you are building, and see who is betting on the same thing.

Free account · 3 free questions · no card

Related cases