Apple's dead car project birthed the AI brains it now runs on
Curated by the Inblix editorial team
It sounds like a tech industry parable: a multibillion-dollar moonshot fails, but leaves behind a piece of technology that reshapes the company. That’s exactly what happened with Apple’s canceled self-driving car. Mark Gurman reports in his latest Power On newsletter that the ambitious project, which never shipped a vehicle, directly led to the development of the Neural Engine — the specialized AI processor that now sits inside every iPhone, iPad, and Mac.
The origin story is a classic case of solving a problem that didn’t exist yet. While designing the autonomous platform, Apple’s engineers hit a wall: they needed massive on-device processing power for real-time AI tasks without torching battery life or phoning home to the cloud. The custom silicon they started sketching never made it into a car. But that work became the foundation for the A11 Bionic chip, which debuted the Neural Engine in the iPhone X. Back then it was a niche component handling FaceID and Animoji. It has since become the quiet workhorse behind every photo you take and every notification that gets intelligently sorted.
Apple’s hardware lead is now its biggest strategic weapon, especially as its software teams scramble to catch up on generative AI. The on-device processing capability lets the company lean hard on privacy as a differentiator — your data stays local because the chips can actually handle the workload. That’s not just marketing; the silicon backs it up. And it’s about to get a lot more serious.
Gurman says Apple is so confident in its direction that it’s leapfrogging the entire M6 family’s high-end variants. No Pro, Max, or Ultra versions of the M6 are coming. Instead, resources are pouring into the M7, slated for the first half of 2027, with a dramatically upgraded Neural Engine. The top-end M7 Ultra is being positioned for servers, packing support for up to 1.5TB of RAM — a spec that reads more like a data center workhorse than something you’d find in a laptop. A failed car project might end up powering Apple’s cloud. That’s one expensive, roundabout pivot.
💡 Key Takeaways
- Apple's Neural Engine, now core to its AI strategy, originated from silicon originally designed for the canceled self-driving car project.
- The company is skipping high-end variants of the M6 chip entirely to accelerate the M7's development, targeting a launch by mid-2027.
- Apple plans to use the M7 Ultra, with support for up to 1.5TB of RAM, as the foundation for a new server product, signaling a serious push into AI cloud infrastructure.
- On-device AI processing remains Apple's primary hardware advantage, enabling privacy-focused features that competitors reliant on cloud processing struggle to match.
Keep reading: See related articles below for more coverage on this topic.
Get smarter about AI
The sharpest AI news, curated daily. Delivered free to your inbox.