MIT students build jet engines with AI copilots in JARVIS Challenge

MIT undergraduates designed, fabricated, and tested small gas-turbine jet engines using large language models as their primary engineering partner in the JARVIS Challenge this past semester, according to MIT News. The Jet-engine AI Research and Validation Intensive Sprint gave 31 students across seven teams four weeks to produce a single-spool engine delivering 50-100 pounds of thrust on Jet-A fuel and completing five 60-second runs. Teams had access to MIT machine shops, commercial tools such as Concepts NREC and ABAQUS, and the new MIT Parley platform, which aggregated frontier LLMs and logged every prompt, model choice, and cost. Corporate sponsors Safran, Voyager Technologies, and Beehive Industries funded essentially unlimited AI access. Professor Zoltán Spakovszky, director of the MIT Gas Turbine Laboratory, said the challenge showed that AI can substantially accelerate safety-critical hardware engineering, but engineering judgment remains the decisive differentiator. Manufacturing, not design or analysis, stayed the fundamental rate-limiting step. Students with little prior turbomachinery experience learned to treat AI as a copilot, knowing when to trust it, when to challenge it, and how to translate its outputs into working hardware. Sponsors described the effort as a genuine experiment in reshaping engineering workflows and a signal that AI-native engineering skills are becoming the future baseline for the workforce.