ROBOTWAR.io
CapabilityB3likely · high confidenceUA · friendly

Harmattan AI demos autonomous Gobi interceptor

Bottom line

US-France firm Harmattan AI demonstrated the Gobi autonomous interceptor, a 350 km/h, 10 km range drone with kinetic kill capability designed for EW-denied environments.

Why it matters

The Gobi's autonomy and EW resistance address a critical gap in NATO air defense against swarms and jammed targets, offering a cost-effective alternative to traditional missiles.

Harmattan AI, a US-France joint venture, demonstrated the Gobi autonomous interceptor at Falcon Peak 26.2. The system features a 350 km/h speed, 10 km range, and container launch capability with kinetic kill. It is designed for fully autonomous trajectory calculation in EW-denied environments. France has ordered 1,000 units. This complements other interceptor developments like the Japanese SkyFall and Cambridge Aerospace’s Skyhammer proposals.

Evidence — 1 independent origin

  • Victory Dronessource Borigin t.me
    Original text (excerpt)

    📹 ❗️ На учениях Falcon Peak 26.2 в Аризоне, организованных Северным командованием ВС США, американо-французская компания Harmattan AI продемонстрировала беспилотный перехватчик Gobi. Технология уже привлекла внимание военных — Франция заказала 1000 таких аппаратов. Ключевые харак

NamedGobiTopicsdronesinterceptorsai warfare

Related findings

← Latest findings · All findings from October 2026

History · version 1 · first published · last substantive update

Cite this finding
Robot War (2026). Harmattan AI demos autonomous Gobi interceptor. Hetzner OSINT, 2 October 2026. https://robotwar.io/osint-hetzner/2026-10-02--harmattan-ai-demos-autonomous-gobi-interceptor/

@misc{robotwar-2026-10-02--harmattan-ai-demos-autonomous-gobi-interceptor,
  title = {Harmattan AI demos autonomous Gobi interceptor},
  author = {Robot War},
  howpublished = {\url{https://robotwar.io/osint-hetzner/2026-10-02--harmattan-ai-demos-autonomous-gobi-interceptor/}},
  year = {2026},
  note = {Hetzner OSINT finding B3}
}

Grades: source reliability A–F from the channel register; information credibility 1–6 from independent origins — reposts count once. The bottom line and summary are written by the model from the cited posts; grades and source roles are rule-based. Methodology 2026-10-15.v6 · Corrections · JSON · RSS · API performance