General Motors spent most of its history with the luxury of a slow clock. A new car could take four to five years from sketch to showroom, and customers mostly waited. That era is over.
The pressure came from China. Chinese automakers now develop new models in roughly 20 months, compared to 32 to 48 months for most Western and Japanese brands. BYD alone received government approval for 38 new or updated car models in a single recent year. Toyota executives reportedly described themselves as shocked after seeing BYD's pace up close during a joint project. GM decided to act.
The response is a bet on virtual engineering. Instead of building physical prototypes early, testing them, discovering problems, and rebuilding, GM's engineers now run thousands of simulated scenarios in software before a car is ever physically assembled. A crash test that once took 15 hours of computing time now takes under one minute. A design change that used to require weeks of back-and-forth between designers and engineers is now visible almost instantly. Physical prototypes still happen, but they arrive later in the process, already refined, which cuts time and cost.
To lead the effort, GM paid $40 million to bring in Sterling Anderson, a former Tesla executive who led both the Model X and the Autopilot self-driving program before co-founding Aurora Innovation, a self-driving truck company. It is the first time since 2001 that GM has hired an outsider to run product development. That detail alone signals how seriously the company is taking the problem.
The GMC Hummer EV is the proof of concept. It went from initial designs to production in about two years, roughly half the normal timeline. GM's stated goal is to make that pace standard across all its vehicles and technology programs.
This matters outside the car industry. The same AI simulation tools are spreading into aerospace, industrial equipment, and consumer goods. The underlying shift is straightforward: anything that used to require a physical test to discover a problem can now, increasingly, be tested in software first. Companies that adopt this approach compress their feedback loops. Companies that do not will find their product cycles looking slow by comparison, regardless of sector.
There is a real caution here too. GM's own president has warned against blindly copying China's approach of moving fast and accepting more mistakes. Speed without quality destroys a brand. The virtual tools are valuable precisely because they are supposed to catch problems earlier, not skip the catching altogether. Whether GM can genuinely match Chinese cycle times without accepting Chinese-style quality tradeoffs remains the open question.
For any business operator watching this: the competitive threat is not just that Chinese manufacturers make cheaper products. It is that they can launch, test, update, and relaunch faster than most Western companies can complete a single development cycle. AI simulation is the Western response. The companies that build these capabilities now, across any industry, will be harder to catch later.