Four astronauts will board a rocket that has only flown once this April to travel to the far side of the moon, 250,000 miles from Earth. It has been 53 years since humans last reached that area.

On March 13, after a flight readiness review, NASA officials held a press conference in response to journalists' questions about the mission's risks, and their answers were impressive not because they were clear, but because they were unusually honest: We know there are risks, but we can't be sure how big they are.

A Question Without a Good Answer

John Honeycutt, chairman of the Artemis II mission management team, was one of the most direct officials at the press conference. He clearly stated that NASA's goal for this mission is to reduce the probability of a flight failure to 1/50, or 2%. But he also admitted that Artemis II has not yet reached that goal.

Super-long-range image of the Orion spacecraft's heat shield. Photo credit: Trevor Malm

"It's probably not half, but it might be closer to half than to 1/50," said Honeycutt, describing the current risk range. This statement carries a lot of information.

Lori Garver, deputy associate administrator for exploration systems development at NASA, questioned the value of quantifying risk itself. She said the agency had completed its risk assessment, but believed these numbers had limited practical value, "Sometimes we are misled into thinking these numbers are telling us something critical."

John Honeycutt, chairman of the NASA Artemis II mission management team, speaks during a press conference at the Kennedy Space Center in Florida on January 16, 2026. Credit: Jim Watson/AFP/Getty Images

These two attitudes together reflect a long-standing structural dilemma in the field of spaceflight risk assessment: numbers require data, but Artemis II has extremely limited reference data. The Space Launch System (SLS) rocket and the Orion spacecraft have only flown once, which was the uncrewed Artemis I mission in 2022. Using a single data point to predict the failure probability of a complex system is statistically unsound.

Honeycutt referred to a broader framework: historically, about half of rockets fail on their first flight. SLS performed well during Artemis I, but making confident judgments about overall reliability based on just one success is clearly not rigorous.

More alarming is the historical lessons. Before the first space shuttle flight in 1981, engineers estimated the probability of crew loss as one in 10,000 to one in 5,000. Looking back, the actual risk of that flight was approximately one in ten to one in twelve, a difference of two orders of magnitude. Honeycutt did not avoid this history; he repeatedly mentioned the Challenger and Columbia disasters, emphasizing that NASA's two worst accidents occurred during the most dynamic phases of missions, not during smooth cruising.

"When we perform liftoff ignition, perigee raise, and trans-lunar injection ignition, those moments will be the nodes where the greatest risk is introduced during the entire mission," he said.

The Astronauts Know and Accept It

The crew members of Artemis II participated virtually in the flight readiness review held at the Kennedy Space Center this week. Credit: NASA/Amber Jane Nottweiler

Captain Reid Wiseman spoke about the dangers of the journey with a calm and clear manner. He told reporters that he had taken his children for a walk and told them where his will and trust documents were, and what would happen to them if something happened to him.

This is not fear, but professional clarity.

Wiseman described some mind-boggling numbers in the mission, such as Mach 39 re-entry speed, a 250,000-mile flight distance, and 5 to 6 million pounds of thrust on the launch pad. "These numbers are beyond comprehension," he said. "There is risk. We don't know what we don't know, so we will learn through the mission."

But he also expressed a strange sense of comfort: "When I get into the Orion, it's like climbing into bed. I feel warm and comfortably squeezed."

Artemis II is currently scheduled for launch on April 1, and NASA has six launch windows in early April. If missed, it will have to wait until the end of the month. The rocket was previously returned to the hangar for repairs due to hydrogen seal leakage and helium issues, and will return to the launchpad next week. The launch team decided not to conduct additional fueling tests, and the next time the rocket is fueled will be on the day of the official launch.

Garver said, "Currently, we have proven that the existing seals are the best we've seen on the SLS." This statement sounds like a guarantee, but it also means that the final verification can only be done at the moment of launch.

Honeycutt finally said that what he most wants to do is "hit that damn entry interface right in the middle to ensure the crew returns safely."

This chart from NASA's Office of Safety and Mission Assurance describes the agency's process for probabilistic risk assessment. Credit: National Aeronautics and Space Administration (NASA)

This statement is perhaps the most honest risk declaration from the entire press conference.

Original: toutiao.com/article/7617048775592280591/

Disclaimer: This article represents the views of the author alone.