The FCC has placed every foreign-made advanced robot on the same national security list it uses for untrusted telecom gear. In public notice DA 26-786, released July 28, 2026, the Commission's Public Safety and Homeland Security Bureau added "all foreign-produced advanced robotic devices" -- and, separately, foreign-produced power inverters -- to the Covered List, acting on two National Security Determinations that a White House-convened interagency body delivered on July 27. The practical effect is immediate: covered equipment cannot receive an FCC equipment authorization, which is the permission slip any radio-emitting device needs before it can be sold in the United States.
Key facts
- What was added: all foreign-produced advanced robotic devices, plus foreign-produced power inverters, in dockets WC 18-89, ET 21-232 and EA 21-233.
- When: the notice was released July 28, 2026, on determinations sent to the FCC on July 27.
- Who acted: the FCC's Public Safety and Homeland Security Bureau, under the Secure and Trusted Communications Networks Act of 2019, on determinations from a White House-convened Executive Branch interagency body.
- Primary source: FCC Public Notice DA 26-786.
The Covered List is a specific legal instrument, not a general blacklist. Congress created it in 2019 to name equipment and services that "pose an unacceptable risk to the national security of the United States or the security and safety of United States persons." Once something lands on it, two existing rules bite: under 47 CFR 2.903(a), covered equipment is prohibited from receiving equipment authorizations, and under 2.911, every applicant must affirmatively certify that its product is not covered. This is why the action needed no new rulemaking, no comment period and no Federal Register wait. It is the enforcement of a rule that has been on the books for six years, pointed at a new category.
The reasoning in the robotics determination reads like a threat model rather than a trade complaint. "The networked capabilities of advanced robotic systems create extensive vulnerabilities and vectors for attacks that can manipulate the data and physical operation of the advanced robotic system," it says, adding that such devices "collect data that could be leveraged by malign actors to surveil Americans, enhance the capabilities of foreign intelligence services, or to remotely commandeer the robots." The two risks the interagency body formally found are supply-chain vulnerability and cybersecurity risk to critical infrastructure. That is the same logic the US applied to network switches and cameras, now applied to machines with cameras, microphones and limbs.
The detail most coverage will lose is the one that most changes the story. The notice states that this newly covered equipment "is identified by place of production, not by entity." No country is named. No manufacturer is named. The parallel power-inverter entry is even blunter, covering inverters produced in a foreign country "regardless of the nationality of the producer." On the text, a US-branded device assembled overseas is inside the category, and ownership is not the test. There is an exit: the Department of War can transmit a determination to the FCC that a specific device or class of devices does not pose the identified risks, at which point the list is updated. For inverters, the Department of Homeland Security can do the same.
What nobody can yet verify is the boundary of the category. The operative definition of "advanced robotic device" lives in Appendix C -- the determination itself -- which the FCC published as a scanned attachment with no machine-readable text. The notice's own prose never mentions any consumer product. So the widely repeated claim that robot vacuums are now banned is an inference drawn from a summary, not something the instrument says, and anyone asserting it should be asked which sentence they are reading. The Commission promises staff guidance to test labs, certification bodies and applicants, with a named contact in the Public Safety and Homeland Security Bureau, which suggests the agency knows the scope question is live.
Why this matters beyond hardware: it is the first US instrument to treat embodied AI as an intrusion surface. The security argument here has an obvious kinship with software-side worries about prompt injection and agent tool exploits -- an AI system that acts in the world can be steered by whoever controls its inputs or its updates. Ground Truth has been tracking the same anxiety in software supply chains, from npm scanning every new package to allied agencies warning that AI-enabled cyber threats are months, not years, away. A robot is a supply chain with wheels.
The honest caveat is that nothing here is settled. No manufacturer has publicly responded. There is no published guidance on how "place of production" will be applied to hardware whose components come from a dozen countries and whose final assembly can move. And the exemption path -- a Department of War determination -- has no precedent, no timeline and no published criteria. Whether this becomes a workable licensing regime or an effective embargo depends entirely on how fast that door opens.
Originally published on Ground Truth, where every claim is checked against the primary source.
Top comments (0)