2026 Mazda CX-50 Recalls, Ratings & Complaints

Incident: October 2, 2026 · Filed: October 21, 2026

The fuel filler door on my 2026 Mazda CX-50 Hybrid premium was defective and would not open due to a sticking fuel tank shut-off valve -- a documented known issue across the 2023 through 2025 CX-50 lineup. The vehicle was delivered brand new with this pre-existing defect without disclosure. The inability to refuel the vehicle creates a direct safety risk of being stranded without fuel with a minor child as a passenger. The defect was eventually confirmed and repaired under warranty however it should have been caught during pre-delivery inspection before the vehicle was delivered to the customer.

Incident: August 28, 2026 · Filed: September 21, 2026

I own a brand-new 2026 Mazda CX-50 Hybrid that is currently under the manufacturer's warranty. I am reporting an intermittent failure of the vehicle to respond normally to a selected driving direction. During the first incident on August 28, 2026, I had shifted the vehicle into Reverse. The vehicle indicated that Reverse was selected, but when I pressed the accelerator, the vehicle did not move or respond as expected. I attempted to change the selected driving direction, but the vehicle was not responding normally to the shift commands.l, again responding as if it were in neutral. I was unable to get the vehicle to operate normally until I turned the vehicle completely off, waited, and restarted it. After restarting, the vehicle began responding normally and was able to engage and move as expected. I captured the first incident on video. The video shows the vehicle's behavior during the event. The same issue occurred again on Sept21, 2026. Again, the vehicle did not respond normally to the selected driving direction and accelerator input. I had to turn the vehicle off and restart it before normal operation returned. This concerns me because the vehicle is intermittently failing to respond and move as commanded. The fact that normal operation returns after the vehicle is shut off and restarted makes the problem particularly difficult to reproduce, but it has now occurred more than once. I have reported the issue to my Mazda dealership and am requesting that the vehicle be inspected and diagnosed under warranty. I have also identified other NHTSA owner complaints involving 2025 and 2026 Mazda CX-50/CX-50 Hybrid vehicles describing similar symptoms, including failure to engage a selected direction, unexpected movement in the opposite direction, lack of vehicle response, and normal operation returning after the vehicle is shut off and restarted.

Incident: May 15, 2026 · Filed: August 3, 2026

On several occasions, moving the gear selector from Reverse to Drive, the transmission has failed to engage Drive, with the dashboard displaying "D" and the vehicle moving in reverse upon throttle application. Fortunately, objects, pets or pedestrians were not present behind the vehicle during the occurrences. The dealership was unable to duplicate the condition and found no software updates or other notices from Mazda addressing the issue.

Incident: July 16, 2026 · Filed: July 24, 2026

I parked this hybrid rental vehicle on a steep street facing uphill and remained in the car with it on for about 15 to 30 minutes. When I put the vehicle back in Drive, released the brake pedal, and pressed the gas pedal to drive away, the car quickly free-rolled backwards. There was no throttle response, and I had to quickly brake to stop the car from its quick backwards descent. Multiple attempts to correct the issue, including various combinations of placing the car back in Park and back in Drive, engaging the electronic parking brake, etc., had no effect. There was no throttle response at all when in Drive, and no ability to control the free-roll of the car after pressing the gas pedal when in Drive without braking. I even tried placing the car in Reverse, and this allowed me to go backwards at a controlled speed instead of free-rolling, but Drive only initiated an uncontrollable backward free-roll. Note that the hybrid vehicle was on, but the engine did not sound like it was running, and I believe it was in electric mode while experiencing this issue. There were no warning lamps, messages, or other symptoms of problems prior to or during the issue. Over the course of these various attempts I rolled backwards several car lengths. Had there been a vehicle or person behind me, there would have been an accident. I eventually placed the car in Park and turned it off and left it there. When I returned about 2 hours later the vehicle behaved as normal, and the issue was never experienced again while I had the rental car. Since this was a Budget rental car, they would need to be contacted about any availability for inspection. I did not report the issue to Budget or to Mazda. This issue concerned me enough to report it directly to NHTSA.

Incident: July 12, 2026 · Filed: July 21, 2026

The contact's daughter owns a 2026 Mazda CX-50. The contact stated that while his daughter was driving 40 MPH and decelerating to turn into a parking lot, the front passenger’s side wheel detached and fell in underneath the vehicle. The vehicle was towed to a tow yard and was then towed to the dealer. The dealer stated that the failure was due to a crash and informed the owner that the repair was not covered under warranty; however, the contact stated that a Police Officer confirmed that there was no damage to the curb or in the vicinity of the vehicle during the failure. No diagnostic information was available. Upon inspection by an insurance adjuster, it was determined that the repair would be covered by the insurance company and would be performed using aftermarket parts. The parts were unavailable, and the vehicle was not repaired. The manufacturer was notified of the failure and assisted in locating a dealer. The failure mileage was approximately 3,506.

Incident: July 14, 2026 · Filed: July 14, 2026

Was braking with traffic and all of a sudden the emergency braking kicked on and brought the car to a stop from about 40 miles per hour with no immediate threat of collision. Brakes were being applied manually and had plenty of room infront of the vehicle and traffic was not stopped. Nearly caused the vehicle behind to collide into me. Very uncomfortable whiplash feeling experienced and my unsecured belongings in my car went flying.

Incident: July 3, 2026 · Filed: July 8, 2026

Component: Air Bags – Knee Airbag / Trim Panel During a collision, on July 3rd with the Mazda CX‑50, the knee airbag deployed and a hard plastic trim panel from the lower dashboard was forcibly ejected. This panel struck my knee and caused a laceration across my kneecap. The injury was not caused by the crash itself but directly by the detached interior component. The trim panel appears to be part of the knee‑airbag assembly and should not detach with enough force or sharpness to cause bodily harm. I experience pain, discomfort, and ongoing issues due to the injury with tingling down my left leg when I get up to walk. I am concerned this represents a design or manufacturing defect in the knee‑airbag trim panel or its breakaway mechanism. Photos of the deployed airbag, the detached part, and the injury are available.

Incident: July 4, 2026 · Filed: July 6, 2026

This report concerns the design of the front seat belt reminder system, which I believe creates a safety hazard. The audible warning begins the instant the engine starts — while the vehicle is stationary in Park — is very loud, and repeats indefinitely until the belt is latched. Federal Standard No. 208 requires only a short warning (up to 8 seconds); this immediate, persistent chime is a voluntary design choice. Other manufacturers (e.g., Honda) meet the same standard while chiming only once the vehicle is in motion, which is sufficient and far less intrusive. The safety problem: because the warning is so aggravating, owners are defeating their own seat belt systems. Owners across the Mazda lineup — CX-5, CX-30, CX-50, CX-90, Mazda3, and Mazda6 — publicly document this on owner forums ([XXX], [XXX], [XXX], [XXX], [XXX], [XXX]) and in TikTok and YouTube tutorials. Common methods include cutting or de-pinning the buckle-latch wire so the car always reads as belted, and installing uncertified seat belt extenders or dummy buckles. The buckle-latch signal feeds the airbag/SRS control module and affects airbag and pretensioner deployment, so these workarounds degrade crash protection. NHTSA's own research notes that reminders annoying enough to be circumvented reduce seat belt use. Mazda has acknowledged the excessive chime volume in Technical Service Bulletin 09-002-24. The system operates as designed — the design itself is the hazard. Symptom: continuous chime and belt warning light at startup while in Park; vehicle available for inspection on request. I request that NHTSA review whether this reminder design induces unsafe circumvention and should be revised. INFORMATION REDACTED PURSUANT TO THE FREEDOM OF INFORMATION ACT (FOIA), 5 U.S.C. 552(B)(6)

Incident: July 4, 2026 · Filed: July 5, 2026

I was backing out of a parking spot in a parking lot that backed up to a 2 lane highway. I placed the car in Reverse and then put the car in Drive, so that I could exit the parking lot to a side road. When I placed the car in Drive and pressed gas, the car went into Reverse (towards the 2 lane highway behind me). I placed the car in park, then back to Drive, it went in Reverse again. We were now in the 2 lane highway with oncoming traffic swerving to avoid us. We decided to put the car in park and turn off the car. Then, we restarted and place the car in Drive. It went into drive and we moved forward into the parking lot, off the highway and then we could exit the parking lot and drive home.

Incident: June 19, 2026 · Filed: June 21, 2026

At approximately 2,900 miles, during the first oil change/service, the factory-installed spin-on oil filter could not be removed normally. The oil filter required approximately one hour of removal effort and would not spin off in a normal manner. When the filter was finally removed, the oil filter threaded adapter/union came out of the engine-side oil filter housing while still seized inside the oil filter. The adapter threads showed visible damage, including flattened, galled, and torn thread areas. The oil filter was seized tightly enough to the adapter that the oil filter had to be cut away from the adapter to separate the parts. The adapter was removed from the filter without adding further damage. The damaged adapter threads were cleaned/chased only as a temporary repair because the vehicle was needed for work and parts were not immediately available over the weekend. A new correct replacement oil filter was test-fitted to the adapter. It threaded fully onto the adapter by hand but had slight binding in the previously damaged thread area. The filter was installed, the engine was started, and the assembly was checked for leaks. No leak was observed during the temporary service check. However, the adapter remains damaged and should be replaced. This appears to be a manufacturing, assembly, or original filter installation issue, because the vehicle had only approximately 2,900 miles and the factory-installed oil filter was seized to the oil filter threaded union. The filter should not be seized to the adapter at this mileage, and the threaded adapter should not be pulled out of the engine-side housing during normal oil filter removal. The concern is that a damaged or improperly threaded oil filter adapter could cause oil filter sealing failure, oil leakage, loss of engine oil pressure, engine damage, or an oil leak onto the roadway. The damaged threaded oil filter adapter/union should be documented as a factory/dealer assembly defect.