NASA Used Claude AI to Plot a Route for Its Mars Rover

NASA's Perseverance rover just did something it's never done before—followed a route mapped out by an AI chatbot.

Between December 8 and 10, Perseverance drove about 400 meters through a rocky section of Mars' Jezero crater following waypoints plotted by Anthropic's Claude. It's the first time NASA's used a large language model to pilot the car-sized robot, and apparently it worked.

It Wasn't as Simple as Asking Claude for Directions

Getting Claude to map a route for Perseverance wasn't just a matter of typing "Hey, can you get this rover from point A to point B?" and letting the AI figure it out.

Routing Perseverance is hard even for humans. "Every rover drive needs to be carefully planned, lest the machine slide, tip, spin its wheels, or get beached," NASA said. Since the rover landed back in 2021, human operators have been painstakingly laying out waypoints—they call it a "breadcrumb trail"—using images taken from space and the rover's onboard cameras.

To get Claude to handle it, NASA had to feed Claude Code, Anthropic's programming agent, years of contextual data from the rover. Only then could the model start writing a route. Claude worked methodically, stringing together waypoints from ten-meter segments, then critiquing and iterating on them.

NASA's Jet Propulsion Laboratory engineers weren't about to blindly trust an AI with a billion-dollar rover, so they double-checked everything. They ran Claude's waypoints through the same simulation they use every day to confirm commands sent to Perseverance. In the end, NASA only had to make "minor changes" to Claude's route. One tweak came because the team had access to ground-level images Claude hadn't seen during planning.

Why This Actually Matters

NASA says using Claude this way will cut route-planning time in half and make the journeys more consistent. Less time on tedious manual planning means more drives, more scientific data, more analysis. In short, in means more learning about Mars.

That's a big deal for NASA right now. Over the summer, the agency lost about 4,000 employees—roughly 20% of its workforce—due to Trump administration cuts. Going into 2026, the president proposed gutting NASA's science budget by nearly half before Congress rejected that plan in early January. Even with funding preserved just below 2025 levels, NASA's being asked to return to the Moon with less than half the workforce it had during Apollo.

Any tool that makes scientists more efficient is going to be welcome.

From Pokémon to Mars

For Anthropic, this is kind of wild. Last spring, Claude couldn't even beat Pokémon Red. In less than a year, the company's models went from struggling with a simple 8-bit Game Boy game to successfully plotting a course for a rover on another planet.

NASA's excited about future collaborations, saying "autonomous AI systems could help probes explore ever more distant parts of the solar system."

So, a chatbot just successfully navigated a rover across Mars. The productivity gains from AI are usually overstated, but in this case, it seems like NASA found an actual use for it that doesn't involve generating terrible stock photos or writing bland marketing copy.

Perseverance has been on Mars since 2021, sending back the first audio recordings from the Red Planet and hitting a bunch of other milestones. Now it can add "first rover to follow AI-generated directions" to the list. Not bad for a robot that's been driving around a crater for nearly five years.

Curious for more stories that keep you informed and entertained? From the latest headlines to everyday insights, YourLifeBuzz has more to explore. Dive into what’s next.