Policy Briefing
Unauthenticated Aviation VHF Voice Communication
July 2026 · 4 pages · US Letter
Last updated July 30, 2026
The question for the record
Does the air traffic control modernization program include any capability to verify that a voice transmitted on an ATC frequency belongs to an actual controller? If not, why not — and what would it cost to add?
In one paragraph
Aviation VHF voice (118.0–136.975 MHz) carries no cryptographic authentication. Pilots comply by recognition. In August 2025 a cloned-controller-voice demonstration was presented on the main track at DEF CON 33. In April 2026, peer-reviewed research at IEEE ICNS reported critical-token F1 = 0.958. Congress has directed more than $32 billion to ATC modernization through 2028 — none of it authenticates the voice on the frequency. This briefing documents the gap. It does not prescribe a technical remedy.
The Problem
Aviation VHF voice communication (118.0–136.975 MHz) carries no cryptographic authentication. Pilots comply with instructions based on voice recognition alone. AI can clone a controller’s voice; language models can generate accurate ATC phraseology. In August 2025, a cloned-voice attack was demonstrated at DEF CON 33. In April 2026, peer-reviewed research at IEEE ICNS confirmed synthetic ATC speech preserves safety-critical tokens with critical-token F1 = 0.958.
The Gap
Congress has appropriated more than $32 billion for ATC modernization, targeting completion by end of 2028. The program replaces 612 radar systems and migrates telecommunications infrastructure. None of it verifies that the voice on the frequency belongs to an actual controller.
The Staffing Factor
The FAA entered 2026 approximately 3,544 certified controllers below its own staffing target. According to NATCA, more than 41% work 10-hour days, six days per week. Fatigue degrades the pattern-recognition that currently constitutes the only defense.
The Legal Gap
Transmitting false ATC instructions is criminal (49 U.S.C. § 46308; 18 U.S.C. § 32). But no federal synthetic-media statute reaches impersonation of a safety-critical operator, and the general aviation-interference statutes predate synthetic voice — they punish the act afterward while requiring no capability to detect or prevent it.
What this briefing does not claim
We do not claim this attack has been carried out. We do not claim any specific airport is vulnerable. We do not claim to know what capabilities the FAA may have deployed and not disclosed. We do not claim the novel’s scenario is a forecast.
We claim that this layer is unauthenticated, that it has been publicly demonstrated, and that the current modernization program does not appear to address it.
Sources
- Vu & Wei (2026), § III — threat model stating the absence of cryptographic authentication and encryption on legacy VHF voice
- Andrew Logan, "Voice Cloning Air Traffic Control," DEF CON 33, August 10, 2025
- Aviation International News, "DEF CON Talk Warns of ATC Voice Cloning 'Nightmare,'" August 22, 2025
- Vu & Wei, "Generating Realistic ATC Voice Communication," IEEE ICNS 2026 (§ V-B, Table I: critical-token F1 = 0.958)
- Reconciliation bill (H.R.1), July 4, 2025
- 49 U.S.C. § 46308; 18 U.S.C. § 32; 47 U.S.C. § 333