{"slug": "dassault-takes-two-ai-algorithms-into-flight-testing-aboard-rafale", "title": "Dassault takes two AI algorithms into flight testing aboard Rafale", "summary": "Dassault Aviation flight-tested two artificial intelligence algorithms aboard the Rafale fighter jet, the French manufacturer announced on September 22, 2026, one developed in-house and the other with cortAIx, the AI accelerator of defense electronics group Thales. Dassault said both sovereign algorithms have reached sufficient maturity to be considered for future Rafale upgrades but did not disclose their functions or which aircraft standard could incorporate them. The tests follow the strategic partnership Dassault and Thales signed on November 18, 2025, to develop controlled and supervised AI for crewed and uncrewed military aircraft.", "body_md": "Dassault Aviation has flight-tested two artificial intelligence (AI) algorithms aboard the Rafale fighter jet.\n\nIn a [statement issued on September 22, 2026](https://www.dassault-aviation.com/en/group/press/press-kits/new-artificial-intelligence-takes-flight-on-dassault-aviations-rafale/) , the French manufacturer said one algorithm had been developed by its own engineers and the other in cooperation with cortAIx, the AI accelerator of defense electronics group Thales. \n\nBoth sovereign algorithms have reached sufficient maturity to be considered for future Rafale upgrades, according to Dassault. The company has not disclosed their functions or identified which aircraft standard could incorporate them.\n\n## **Dassault and Thales move from AI partnership to flight testing** \n\nThe tests are part of Dassault’s effort to introduce controlled and supervised AI into the cockpit, with the technology intended to support the human crew.\n\nThe manufacturer identified several technical challenges involved in bringing AI aboard combat aircraft, including obtaining reliable operational data, combining expertise from different disciplines, and adapting algorithms to the limited computing resources available on embedded systems.\n\nThe flight tests follow the [strategic partnership signed by Dassault and Thales](https://www.aerotime.aero/articles/dassault-thales-controlled-defense-ai-partnership) on November 18, 2025, to develop controlled and supervised AI for crewed and uncrewed military aircraft. \n\nThe agreement covers applications spanning observation, situation analysis, decision-making, mission planning, and control.\n\nThales has already been developing AI functions for the Rafale’s TALIOS targeting pod. In November 2024, the company said an [AI-assisted capability designed to detect and pre-select objects of interest](https://www.aerotime.aero/articles/thales-enhances-rafale-reconnaissance-with-ai-driven-talios-pod-upgrade) for the pilot was expected to enter service with the F4.3 standard in 2026. \n\n## **Harmattan AI partnership targets Rafale F5 and uncrewed aircraft** \n\nDassault is pursuing a parallel effort with French defense startup Harmattan AI. On January 12, 2026, the manufacturer announced that it was leading a $200 million Series B funding round for the company as part of a partnership to develop [embedded AI for the Rafale F5 and Dassault’s uncrewed combat air system](https://www.aerotime.aero/articles/dassault-harmattan-ai-controlled-autonomy-rafale-f5-ucas). \n\nThe collaboration focuses particularly on the control of uncrewed aircraft, a capability central to the Rafale’s future development.\n\nThe two companies demonstrated another aspect of their cooperation on July 13, 2026. During a flight test, a drone equipped with their jointly developed NAMIB electronic warfare payload detected and geolocated a radar several dozen kilometers away, then transmitted its position to a Rafale F4, which simulated a strike against the target.\n\n## **Rafale F5 prepares for greater AI integration** \n\nThe latest flight tests come as development of the Rafale F5 gathers pace.\n\nOn September 10, 2026, France’s defense procurement agency, the Direction générale de l’armement (DGA), announced that it had awarded preliminary [development contracts for several major components of the next Rafale standard](https://www.aerotime.aero/articles/france-rafale-f5-radar-engine-ew-development) . \n\nAmong them is the Thales RBE2-XG radar, which will feature increased computing capacity to support greater AI integration, [according to the DGA](https://www.defense.gouv.fr/dga/actualites/dga-commande-premiers-travaux-amont-du-prochain-lancement-realisation-du-standard-f5-du-rafale) . \n\nThe agency plans to launch the F5 into its realization phase by the end of 2026, with qualification targeted for around 2033.\n\nDassault is not the only European fighter manufacturer taking AI into flight testing. In May 2025, Saab and German defense startup Helsing [tested the Centaur AI agent aboard a Gripen E](https://www.aerotime.aero/articles/saab-gripen-e-ai-flight-tests). During simulated beyond-visual-range combat, Centaur autonomously maneuvered the fighter and cued the pilot to fire.", "url": "https://wpnews.pro/news/dassault-takes-two-ai-algorithms-into-flight-testing-aboard-rafale", "canonical_source": "https://www.aerotime.aero/articles/dassault-rafale-ai-flight-tests-future-upgrades", "published_at": "2026-09-22 10:58:32+00:00", "updated_at": "2026-09-22 11:53:53.248271+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-policy"], "entities": ["Dassault Aviation", "Rafale", "Thales", "cortAIx", "Harmattan AI", "Direction générale de l'armement", "Saab", "Helsing"], "alternates": {"html": "https://wpnews.pro/news/dassault-takes-two-ai-algorithms-into-flight-testing-aboard-rafale", "markdown": "https://wpnews.pro/news/dassault-takes-two-ai-algorithms-into-flight-testing-aboard-rafale.md", "text": "https://wpnews.pro/news/dassault-takes-two-ai-algorithms-into-flight-testing-aboard-rafale.txt", "jsonld": "https://wpnews.pro/news/dassault-takes-two-ai-algorithms-into-flight-testing-aboard-rafale.jsonld"}}