First a confession… I’ve been predicting Full Self-Driving cars would be a reality “this year” for about a decade, so you might want to take my predictions with a grain of salt. 😂😂 However, I recently got my hands on Tesla’s Full Self Driving version 12, marking the first iteration that leverages machine learning and neural networks, akin to AI systems like ChatGPT. This new approach eschews traditional programming for a model where the system learns by observing the world’s best drivers (with consent) and emulates their driving styles. This is a huge advancement for many reasons. Thanks for reading Year Of The Opposite - Travis Stoliker's Substack! Subscribe for free to receive new posts and support my work. * By using end to end neural networks (Think AI), instead of software developers needing to code millions of lines of descriptive computer code, (A stop sign looks like this… This is what you do when you see a stop sign… ETC) with the new system, Tesla was able to throw away more than 300,000 lines of explicit C++ code. Now, quite literally, the computer is writing it’s own code! * Because the system is now learning by watching and mimicking the best drivers, it drives much more smoothly. The drive feels more natural. More like a human is driving. * Development is much faster now. Instead of software developers needing to explicitly program every line of code, the computer system is learning and coding by itself! * The computer learns by observing, not being told what to do. For instance, in the past when the programming was all done by humans, a human would have to program into the system how to avoid potholes. But now since the system is just observing what great drivers do and mimicking their behavior, the system has taught itself how to identify potholes and what to do to avoid them. It’s not perfect yet. But it’s getting pretty good. When we returned from our trip to North Carolina it was our first long car ride using Full Self Driving Version 12. With the old software, the system would drive us about 90% of the drive. But this drive was MUCH better! The system handled everything from highways to parking lots seamlessly. The only manual tasks I had to perform were deciding on parking specifics, like whether we were stopping for a drive-thru or parking to go inside the restaurant. Laken, who has been skeptical of the jerky nature of the Full Self Driving (Beta) we’ve had for four years, was impressed with the new system. The latest version drives smoothly, mirroring human-like driving precision. It’s quite the game-changer. With 3 million Teslas already equipped with the necessary hardware out of the 6 million produced, the implications are enormous. As these cars and the 2 million new Teslas produced annually start performing autonomous Uber rides, a lot of cool things are going to happen: * A significant number of lives will be saved. * It will compel other car manufacturers to license Tesla’s software, especially after the recent adoption of the Tesla charging standard by most car manufacturers. * Most cars, which currently sit idle and parked 95% of the time, represent a huge waste of resources. Now, those same cars could be out driving autonomously, performing Uber rides and generating income. For many Americans, a car is the second largest expense they have; this shift has the potential to turn it into a profit-making asset. To say I’m excited is an understatement. With the flip of a switch, 3 million cars on the road today are going to turn into Robots on 4 wheels. It’s going to save a lot of lives since car accidents are the number one cause of death for young Americans. And it’s going to make transportation cleaner, safer, and cheaper. Mark my words, I predict that we will look back and say that April 2024 was the date that Humans Invented Self Driving Transportation. 1 Month Free for Paid Subscribers A quick housekeeping update. You proba