Reaching our nearest star system has long remained an exercise in theoretical ambition, stalled by immense physical distance and prohibitive launch costs. Alpha Centauri sits 4.4 light-years away—roughly 25 trillion miles from Earth. NASA's Voyager 1 probe has been travelling since 1977 and has covered less than 1% of that distance, meaning a conventional journey would take over 70,000 years. Previous private initiatives, such as the Breakthrough Starshot mission announced by billionaire tech investor Yuri Milner in 2016, pledged $100 million toward a laser-propelled proof of concept to reach the system in 20 years, yet nothing has launched a decade later.
Rethinking the Mission Envelope
A nonprofit organization called the Fermi Explorer Mission has taken an entirely different computational approach, targeting a spacecraft launch to Alpha Centauri by the end of 2029. Instead of designing around a human lifetime, the team aims to deliver a cargo payload weighing at least one kilogram—carrying artistic and scientific payloads, messages, and a copy of NASA's 1977 Golden Record—on a projected budget of just $15 million funded by individual private donors. For a year, the team struggled to design a small solar-powered craft that could generate sufficient propulsion without ballooning in weight.
Philip Johnston, cofounder and president of the Fermi Explorer Mission, discussed this engineering dead end on a podcast hosted by Alex Wissner-Gross, a physicist and cofounder of the newly launched AI physics research lab Physical Superintelligence (PSI). PSI, which launched with $58 million in funding led by the Bill Gates-founded investment group Breakthrough Energy, deployed its open-source system called Get Physics Done to tackle the problem.
Computational Discovery in Complex Physics
The AI system synthesized orbital maneuvers to generate a route the human engineering team had missed, navigating multi-dimensional physical parameter spaces to balance solar radiation pressure and orbital mechanics. This orbital strategy keeps the total mass light and viable under strict budget limits, showing how automated reasoning can transition from text generation to hard physics and aerospace optimization.
Current Boundaries of Automated Research
If the Fermi Explorer spacecraft launches on schedule in 2029, it could take up to 80,000 years to reach Alpha Centauri. The project will not deliver immediate interstellar exploration, but that is not where its commercial utility lies. What this mission demonstrates is a concrete method for using AI optimization models to explore complex parameter spaces that human engineering teams overlook—a computational paradigm with immediate application in aerospace, energy grid management, and complex materials modeling.