On March 10, Ethiopian Airlines Flight 302, a Boeing 737 Max 8 aircraft, took off from Addis Ababa in Ethiopia and almost immediately suffered flight control issues. Within five minutes of takeoff, the aircraft went down, claiming the lives of all 157 people on board. Tragic as this incident was, it didn’t take long before investigators and journalists began linking it to another crash only five months earlier. In that incident, a Lion Air-flagged Boeing 737 Max 8 crashed into the Java Sea near Indonesia shortly after takeoff, killing all 189 people aboard.

Both aircraft were Boeing 737 Max 8s. Both had been delivered only months before their deadly crashes. And now, it appears both planes shared issues with the aircraft’s flight angle just before crashing — an issue that seems to be systemic to the aircraft’s flight management software.

According to an analysis of the cockpit recorder of the Ethiopian Airlines crash, the plane’s “angle of attack” was very similar to the data recorded from the Lion Air crash. Experts now think this flight angle issue may have been caused by a faulty angle-of-attack sensor connected to an automated system that dips the nose of the aircraft downward when the computer thinks there’s a risk of the flight angle causing a stall.

“If that’s the case, that does raise the possibility that there is a similar occurrence between the Lion Air and Ethiopian Airlines accidents,” said Clint Balog, a Montana-based professor at Embry-Riddle Aeronautical University. He also pointed out, however, that the investigation is still in its infancy.

The flight control system in question, called MCAS (Maneuvering Characteristics Augmentation System), has become the focus of investigators who point out how Boeing apparently rushed the 737 MAX line into production with assistance from the Federal Aviation Administration (FAA). In the rush to field the new platform, the FAA chose to delegate safety assessment pertaining to the aircraft, including for the flight system in question, to Boeing itself. According to the accounts of engineers and safety officials involved, that decision may have led to Boeing understating potential risks, not relaying information effectively to pilots and operators, and even downplaying the level of control MCAS has over the aircraft equipped with it.

“There wasn’t a complete and proper review of the documents,” a former engineer told the Seattle Times. “Review was rushed to reach certain certification dates.”

Boeing officials haven’t disputed the idea that their MCAS may have caused these crashes, but they have questioned the roles crewmembers may have played in allowing them to occur. According to Boeing, pilots should be aware when their horizontal stabilizers are not mirroring their controls. If the pilots in both of these crashes followed the checklist procedures for “stabilizer runaway,” Boeing contends, even a malfunction of the MCAS could be arrested in time to save the aircraft by following those procedures.

Boeing’s position is bolstered by reports of other aircraft experiencing similar issues and doing exactly that: resolving the issue by following the prescribed procedures and continuing on their way. However, pilots and safety officials have taken issue with Boeing’s explanation, claiming that these failures don’t exhibit the same symptoms of stabilizer runaway. They also added that the MCAS disabled a function that would interrupt the nose down movement by pulling back on the column.

According to an FAA safety engineer’s appraisal of the Lion Air crash, Boeing assumed that “the pilots would recognize what was happening as a runaway and cut off the switches.” He went on to opine that, “The assumptions in here are incorrect. The human factors were not properly evaluated.”