
Someone or something in the cockpit of a China Eastern Boeing 737 switched off the jetliner’s engines causing the plane to crash short of its destination in Guangzhou, China in 2022. This comes from an analysis of the plane’s flight recorder by the U.S. National Transportation Safety Board and it quickly got tied to the still unexplained crash of Air India Flight 171 nearly a year ago. But are the two related? And should air travelers be worried about the people flying their airplane? Let’s take a sensible look at the facts.
The unusual release of what is normally confidential investigation information from the China Eastern disaster in March 2022, was reported on May 4th. Shortly after the accident, however, the aviation industry website, Leeham News and Analysis hosted a healthy debate among its readers and the Wall Street Journal ran a story quoting an unidentified source suggesting the disaster was an intentional act.
These anonymous claims and industry speculation got a credibility boost with the publication last week of the data the NTSB experts coaxed from the black boxes. The one paragraph analysis follows an 18 page description of the extraordinary lengths engineers went to extracting data from the voice recorders’ components. (You won’t read it here so for more on epoxy bonds, deformed pins and chip cracks, see pages 4-22)

According to the NTSB’s analysis, the fuel switches for both engines moved from run to cutoff at 29-thousand feet, as Flight 5735 was enroute from Kunming to Guangzhou. The engines lost speed so quickly, they could no longer provide power to the airplane’s systems including the flight data recorder, which stopped functioning as the plane passed through through 26-thousand feet. The cockpit voice recorder had battery backup and continued to function.
China Eastern 5735’s descent was stunningly swift. It plunged five and a half miles in two minutes. A video of the plane in its last seconds shows it traveling nearly vertical.

The NTSB assessment was that, “The fuel switches on both engines moved from the run position to the cutoff position.” There is no statement about who moved the switches or whether the cockpit voice recorder contained any information to explain what happened.
A few things about the NTSB’s work on Flight 5735 are curious.
Though the work was done at the NTSB’s lab in Washington, DC, and the safety board is entitled under international agreement to participate in read-outs of data and flight recorders, all the data extracted was sent to the Civil Aviation Administration of China, nothing was retained by the NTSB.
“As an accredited representative to the CAAC’s investigation, the NTSB assisted by downloading the FDR and CVR,” said Peter Knudson, an NTSB spokesman. “The CVR and audio file were in the possession and control of the CAAC, which is leading the investigation.”
Not only was NTSB left in the dark about the contents of the CVR they were so critical in extracting, the American companies, Boeing and Honeywell Aerospace, maker of the recorders were excluded from participating in the data recovery group. Their role, Knudson said was in providing “technical assistance in extracting the audio from the CVR. They did not listen to the audio file.”
The third curious thing is that four years have passed since the crash. No accident report has been issued. Chinese officials are mum.
In my inbox, however, several people are comparing the China Eastern disaster to the June 12, 2025 crash of an Air India 787 in Ahmedabad. Only a highly superficial view could lead to that conclusion
Air India Flight 171 departed for London on June 12, 2025, but 30 seconds later flew into a building killing all but one of the 242 people on board and 19 people on the ground.

In its preliminary report, the Air Accident Investigation Bureau of India reported during the short flight, the engine fuel switches “transitioned” to the cutoff position. This triggered countless news stories that pilot suicide was the likely cause. That’s where the parallel to China Eastern begins and ends.
A few intrepid reporters have pointed out that trouble was brewing on the Air India Boeing 787 registration VT-ANB before fuel to the engines stopped flowing. The most obvious issue is clear as day in many videos of the flight lifting off.

Even as the plane lifts off the runway, a small fan designed to provide power in emergencies can be seen protruding from below the fuselage. Its three blades would be pinwheeling in the airstream.
This tiny device, known as a ram air turbine or RAT is used on many modern airliners including the Boeing 787. To put a grim spin on this tiny fan, it is an instrument of last resort. When the engines have lost power, or the plane’s main electric or hydraulic systems fail, the fan harnesses the dwindling forward momentum generating enough power to give pilots basic flight control and instruments.
I mean, this is what it is designed to do. But it takes time to begin working and only helps if the plane is flying high enough. As close to the ground as was Flight 171, there’s not much the RAT could do.

The RAT at takeoff is not only useless it should not have happened. Something had to be terribly amiss on the airplane for it to deploy and no explanation has come from Indian investigators.
“Engineers say that points directly to a system failure not a pilot error,” writes Rachel Chitra, a journalist who has been reporting on this accident since it happened for publications in India. The timing of the RAT deployment is clear evidence “the aircraft was already behaving as if it had lost electrical power before the engines actually failed,” Chitra writes.
The question is why.
When air accidents happen, many of my colleagues, air safety specialists, aviation journalists and aviation geeks are fond of saying, “Don’t speculate, let the investigation run it course.” That makes no sense. When knowledgeable folks outside of the official investigation use their own eyes, experience and logic they enhance the process and hold investigators to a credible standard.
In my book, The Crash Detectives, an entire section is devoted to arm chair investigators who exposed flawed or deliberately misleading official reports. (For one mind-blowing story, read about the man who solved the mysterious crash of Air New Zealand Flight 901.)

Countless words have been written about the Air India crash, many of them suggesting – or worse concluding – deliberate pilot action brought down the plane in some sort of murder suicide. A Wall Street Journal article in November drew the ridiculous conclusion that “The crash increasingly appeared to be the latest in a grim string of crashes caused by ‘pilot suicide and mass murder,” according to American government and industry officials.”
Pinning the opinion on “officials” does not relieve the paper of the obligation to question whether there really has been a “grim string of crashes,” so let’s count them.
In the 50 years between 1975 and the time of Air India crash in 2025 there have been 906 fatal airliner crashes. Six of them, or two thirds of one percent of all airline disasters were believed to be the result of deliberate pilot action.

I can’t write much about the China Eastern disaster. Little information is emerging which is one reason the NTSB’s release of a scant few details has led to a new round of speculation that airline pilots are on a murder spree.
It would be a wrong to jump to that conclusion, not just because there’s not enough to go on – as in the China Eastern accident – but because there’s a significant anomaly in the Air India crash that anyone with eyes can see. And despite the loss of 260 people, and the millions of words written about the disaster, that clue isn’t getting the attention it should.








As an aviation safety SME we are taught to avoid the speculation. The data will tell the story, this is true in 99% of the accidents. In my career I’ve worked with the CAAC and India DGCA. No surprise that both are mum about these accidents. Both the 737 and the 787 fuel cutoff switches must be physically actioned by the crew for both engine start and engine power cutoff. Could an errant signal shutoff the fuel? On the 787 possibly, but not the 737.
The term “fuel control switches” is misleading. The switch in the cockpit is only one of the inputs to the TCMA and FADEC. TCMA stands for Thrust Control Malfunction Accommodation. It’s a software system that controls the Full Authority Digital Engine Control (FADEC) that monitors engine thrust and can intervene to prevent or mitigate potential thrust-related issues, particularly during landing or when an engine is not responding as commanded. There is a single TCMA and each engine has its own FADEC.
The fuel flow is actually controlled by the fuel control valves in the engine. That is controlled by the TCMA/FADEC. It is important to understand that, if the TCMA/FADEC or its sensors malfunction, the fuel can be switched off without any action by the pilots.
Chris says “Both the 737 and the 787 fuel cutoff switches must be physically actioned by the crew for both engine start and engine power cutoff. Could an errant signal shutoff the fuel?” I understand this to mean a signal independent of the switches’ physical position. Correct?
Why, then, would this be possible on the 787 but not the 737?
It is my understanding, Rod, that the 787 fuel cutoff switches in the cockpit are one of TWO ways that fuel can be terminated to the engines. There is secondary control to allow the TCMA to activate. As you probably know, when certain unsafe conditions are exhibited, the thrust control malfunction accommodation kicks in and shuts down the engines. This is a digital, not a switch activation. This would also be noted on the DFDR so investigators should likely know about this, including whether the fuel flow stopped by cockpit mechanism or by digital control.