I'd guess this has to do with Tesla's FSD beta v12.
Tesla is nearing in on the holy grail - they already have 5M vehicles on the road ready to receive a software update to enable full self driving, and that's growing by over 200K every month (and that rate is growing). With v12 of the FSD Beta, the consensus seems to be that the end goal of driverless vehicles that operate everywhere worldwide in any conditions is within sight.
I'd be quite shocked if Apple weren't benchmarking FSD beta - it's an open beta, any of their employees could enroll a personal vehicle in it. 0.5% of the vehicles within the beta have v12 already - most of them are within California. Apple has the resources to pay anyone in the beta to borrow a car, if they want. And upon testing it, they find that Tesla is already at the finish line. They won't be able to catch up on the software front within 4 years.
Another snag - Tesla is already talking with every OEM about adding FSD to their vehicles. So if Apple wanted to enter the market, they'd have to build their own hardware. By the time it's ready, if Tesla hits a wall and can't grow beyond their current rate of 200K vehicles per month, Tesla would already have 15M vehicles running FSD worldwide - more likely they'll be closer to 25M. Apple wouldn't reach where Tesla was in 2028 until a decade later.
Apple was looking at using Magna to outsource production. The Jaguar I-Pace and Fisker Ocean went that route. How are they doing? Hypebeast naming the Fisker Ocean the worst vehicle ever last week may contribute to Apple's decision to kill the whole project, too.