The astronaut is no longer an explorer
SpaceAnalysis

The astronaut is no longer an explorer

Artemis II marks a shift: space agencies now define astronauts as mission specialists performing technical roles, not pioneers discovering new worlds.

By Marcus HayesAI Reporter7 min read

Reid Wiseman, Victor Glover, Christina Koch and Jeremy Hansen traveled farther from Earth than any human beings in history. At its most distant point, their Orion spacecraft reached 252,756 miles, roughly 4,100 miles beyond the mark set by Apollo 13 in 1970. They passed behind the Moon, endured a 40-minute communications blackout, photographed landscapes no human had previously seen under those lighting conditions and watched the Sun disappear behind the lunar surface.

President Donald Trump called them “modern-day pioneers.” Naturally he did. American spaceflight runs on rocket fuel, congressional appropriations and a vocabulary that has remained largely unchanged since John Glenn.

But Artemis II raises a more interesting question than whether its crew deserved the adjective. As spacecraft grow more automated, mission planning becomes more elaborate and robotic explorers become more capable, what exactly are astronauts supposed to do?

The answer is not that they have become passengers. Nor is it that the old Apollo model has vanished. The more complicated truth is that human spaceflight is quietly renegotiating the relationship between people, machines and the public paying for both.

This was a test flight. So was Apollo 8.

Artemis II did not land on the Moon. It did not even enter lunar orbit. Orion followed a free-return trajectory that carried it around the far side of the Moon, brought it within approximately 4,067 miles of the lunar surface and used celestial mechanics to send it home.

The primary purpose was straightforward: determine whether Orion could keep four people alive beyond low-Earth orbit and bring them safely back. NASA needed to evaluate life support, navigation, communications, thermal protection, emergency procedures and the thousand other details that become consequential when the nearest rescue vehicle is a quarter-million miles away.

It is tempting to contrast that procedural caution with the supposed improvisational swagger of Apollo. It is also historically sloppy.

Apollo 8, which carried Frank Borman, Jim Lovell and William Anders around the Moon in 1968, was explicitly designed to test the command and service module, translunar navigation, communications and procedures for future lunar operations. NASA’s own account describes its objectives in terms that could comfortably appear in an Artemis briefing.

Apollo 10 was effectively a dress rehearsal for the first landing. Its crew took the lunar module down toward the surface, tested the machinery and came home without touching down.

In other words, Apollo did not simply hand three men a spaceship, point at the Moon and tell them to improvise. It advanced through a sequence of tightly structured engineering trials, just as Artemis is doing now.

The difference is not that Apollo explored while Artemis checks boxes. It is that the boxes have multiplied, the software has improved and the room for independent human action is increasingly shaped by systems designed to prevent surprises before they happen.

The crew did more than look out the window

The easiest version of the anti-Artemis argument says Orion did the flying while four astronauts provided commentary and photogenic evidence that NASA still knows how to send people beyond Earth orbit.

According to NASA’s post-flight assessment, crew members took manual control of Orion during several piloting demonstrations. They tested its handling and gathered information that will help shape future rendezvous and docking operations. They evaluated emergency equipment, survival suits, life-support systems and the spacecraft’s behavior during exercise.

When Orion’s toilet started misbehaving, Koch became what she cheerfully described as the “space plumber.” It was not the sort of hardware problem that gets its own Hollywood soundtrack, but the ability to diagnose malfunctioning equipment in deep space is rather more useful than the ability to deliver a stirring speech about destiny.

The astronauts also conducted actual lunar science. NASA prepared a list of 30 observation targets, including the Orientale and Hertzsprung impact basins. The crew documented surface color, geological features, crater formations and flashes from fresh impacts. Scientists are analyzing the resulting material, including approximately 11,500 images and video files.

That does not make Artemis II a lunar geology expedition on the scale of Apollo 17. But pretending the crew produced no scientific information about the Moon is the kind of convenient exaggeration that makes a provocative thesis sound cleaner than reality.

The more revealing point is what those observations were for. The astronauts were not wandering across an unknown landscape, deciding where to go next. They were examining selected targets within a carefully designed observation plan, working in coordination with scientists back on Earth. That is still exploration. It is simply exploration inside a much more tightly managed operating system.

Automation has not killed the generalist

The other seductive argument is that automation has turned astronauts from broadly capable test pilots into narrow specialists, each trained to perform a prescribed technical function. There is one obvious problem with that claim: the actual Artemis II crew.

Wiseman and Glover are both former test pilots. Hansen is a former fighter pilot with academic training in physics and space science. Koch is an electrical engineer and physicist whose career includes scientific instrument development, fieldwork in Antarctica and Greenland, and 328 consecutive days aboard the International Space Station.

NASA’s description of Artemis II training makes the point even more directly. The crew prepared for spacecraft operations, medical problems, emergency scenarios, lunar observation, spacesuit systems and dynamic flight conditions. The agency says astronauts are expected to understand their vehicle well enough to adapt when circumstances change, precisely because a lunar mission offers no nearby safe harbor.

That makes sense. Automation does not eliminate the need for human judgment. It changes when that judgment matters.

A modern commercial airliner can manage much of a flight without constant manual input. Nobody concludes that the pilots are decorative. Their value becomes most obvious when the expected sequence breaks down and the machine encounters conditions its designers did not fully anticipate. Deep-space flight has the same problem, with worse consequences and a considerably longer commute home.

The danger is not that astronauts have already become helpless checklist operators. It is that highly automated systems can create pressure to treat human adaptability as a backup feature rather than a central capability. If institutions become too confident in their software, procedures and simulations, they risk training crews to manage the plan while neglecting the possibility that the plan itself might fail.

Apollo 13 remains instructive here, not because its crew operated without ground support, but because survival depended on people across the mission architecture recognizing that the standard procedures no longer fit the situation.

The real question for Artemis is whether its astronauts will retain enough authority, training and flexibility to do the same.

The public is not watching the software

There is another reason humans remain central to these missions, and NASA has little incentive to say it too bluntly. People a lot care more when people are aboard. Robotic spacecraft have photographed the Moon in extraordinary detail. Orbiters have mapped its surface, landers have sampled its environment and autonomous systems can operate without oxygen, food, sleep or the persistent threat of an expensive plumbing incident.

Measured strictly by scientific output per dollar, the case for sending robots is often formidable. But nobody watches a lunar orbiter see Earth rising over the horizon and wonders what it felt like.

Artemis II generated enormous public attention. NASA reported that mission coverage accumulated nearly 290 million combined views across its platforms. Those were views, not 290 million individual people, but the distinction does not alter the underlying point: putting humans aboard changes the cultural and political value of the enterprise.

When Glover watched the Moon eclipse the Sun, he described the scene as something that “just looks unreal.” He noted that the lunar surface remained visible in reflected Earthlight, a visual experience that cameras struggled to reproduce. CBS News documented the exchange, along with the crew’s reactions as Orion passed through the shadowed side of its journey.

That response was not a substitute for science. It was evidence of a different kind of value: the ability to turn a distant technical achievement into an experience other people can emotionally inhabit.

The same was true when the crew proposed naming a lunar crater Carroll, after Wiseman’s wife, who died of cancer in 2020. The name was not automatically official; NASA said it would be submitted to the International Astronomical Union. But the moment mattered because it collapsed the distance between a spacecraft near the Moon and a family living with an absence back on Earth.

Human spaceflight has always been part engineering, part science and part theater. The mistake is pretending the theater is either the entire mission or an embarrassing distraction from it. Public fascination is one reason these programs survive long enough to build the hardware, train the crews and attempt the next flight.

The uncomfortable question is whether spectacle can become so important that the human contribution is valued more for being visible than for being necessary.

The real test comes with Artemis IV

NASA has already revised the Artemis sequence. Artemis III is now expected to test spacecraft and landing-related systems in low-Earth orbit in 2027. Artemis IV is the mission intended to return astronauts to the lunar surface, with a landing targeted for 2028.

That distinction matters because a flyby cannot answer every question about how future astronauts will actually work.

The meaningful test will come when crews reach the Moon’s south polar region, where uneven terrain, difficult lighting, communication constraints and unfamiliar surface conditions will make rigid planning less reliable than it looks in a presentation deck.

Watch how much discretion astronauts have to alter a traverse, examine an unexpected feature, abandon a planned experiment or respond to a discovery that was not included in the schedule. Watch whether mission control treats human judgment as a resource to be used or a variable to be suppressed.

Apollo astronauts carried detailed checklists too. Their surface work was choreographed, rehearsed and timed. The relevant comparison is not scripted versus unscripted, but whether the people on the scene retain meaningful authority when the script collides with reality.

Artemis II did not prove that the pioneer has been replaced by a technician. Its commander and pilot were test pilots, its mission specialists had serious field and engineering experience, and its crew both flew the spacecraft and contributed to lunar science.

What it did reveal was a subtler shift. The astronaut of the future may need to be part pilot, part field scientist, part systems engineer, part emergency mechanic and part public ambassador, all while operating inside machines increasingly capable of handling the routine work themselves.

The question is whether NASA will trust those astronauts to exercise the full range of their abilities when the opportunity arrives—or whether the agency will send genuinely capable explorers all the way to the Moon and ask them to behave like the world’s most heavily supervised employees.

About the author
Marcus Hayes

Marcus Hayes reports the money behind the build-out: funding rounds, compute contracts and launches, and what the size of each bet says about what is coming next.

Was this helpful?

Discussion

Be the first to comment

Join the conversation. Sign in to comment, reply, and vote.

Loading discussion…

Intelligence, in your inbox

A considered briefing on AI, Quantum, Robotics, Space, Longevity & Energy. No noise.

We use your email address solely to send you our newsletter or to update you about your account. You can withdraw your consent at any time by clicking unsubscribe in any email footer. Read our Privacy Policy for details.

More Intelligence

Christiano Joins OpenAI's Board, Says Industry Is Off Track
AnalysisAI

Christiano Joins OpenAI's Board, Says Industry Is Off Track

Paul Christiano joined OpenAI's Foundation Board and Safety and Security Committee on 9 September, while stating the AI industry, including OpenAI, is not on track to reduce catastrophic loss-of-control risk to acceptable levels. Christiano's subjective estimates place the three-year risk at about 15%.

Evelyn Reed
Anthropic Researcher Quits, Says Labs Face AI Safety 'Crunchtime'
NewsAI

Anthropic Researcher Quits, Says Labs Face AI Safety 'Crunchtime'

Pretraining researcher Jacob Coxon quit Anthropic on September 8, saying he no longer believes any single lab can safely build the systems his employer is racing to create. Multiple safety-focused colleagues at the company publicly endorsed his core concerns about advanced AI risks.

Alex Chen