{
  "id": 26889,
  "title": "Can AI build a jet engine? JARVIS Challenge tests role of AI copilots in tough-tech engineering",
  "url": "https://urgent.news/2026/07/14/can-ai-build-a-jet-engine-jarvis-challenge-tests-role-of-ai-copilots",
  "topic": "ai",
  "section": "AI",
  "published": "2026-07-14T18:00:00.000Z",
  "source": {
    "name": "MIT News AI",
    "slug": "mit-news-ai",
    "url": "https://news.mit.edu/2026/can-ai-build-jet-engine-jarvis-challenge-tests-ai-copilots-in-tough-tech-engineering-0714"
  },
  "original_language": "en",
  "account": "Artificial intelligence has revolutionized software engineering, but its impact on designing and manufacturing complex physical systems, like jet engines, remains uncertain. The JARVIS Challenge, a recent competition at MIT, strove to answer this question by giving undergraduate students four weeks to design, fabricate, assemble, and test a small gas turbine aero engine. The objective was to build a small jet engine capable of producing 50-100 pounds of thrust and completing five 60-second runs using Jet-A fuel.\n\nThe challenge involved 31 students from various departments at MIT, organized into seven teams. The teams had complete freedom in designing, choosing materials, and fabricating their engines. They had access to MIT's machine shops, manufacturing vendors, commercial software, and test rigs. Furthermore, the students had access to MIT Parley, an interface for large language models (LLMs) that helped them use AI in their projects.\n\nWhile AI proved helpful in summarizing information, teaching AI usage, sourcing vendors, creating Excel sheets, answering specific questions, finding references, and creating comparative analyses, it had limitations too. AI struggled with design creation, often producing hallucinations and sycophancy, which slowed down the teams and reduced their confidence in the design. By the end of the challenge, many teams had to rely on their engineering expertise to overcome these limitations. Professor Spakovszky, the director of the MIT Gas Turbine Laboratory, noted that while AI could substantially accelerate safety-critical hardware engineering, it was engineering judgment, not AI, that made the decisive difference.",
  "summary": "MIT students designed, built, and tested a jet engine with AI copilots, assessing AI’s usefulness in developing high-performance aerospace systems.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}