The Age of AI Helped Along by Curious Garage-Based Innovators
Maybe it was just me not paying attention to the computer universe that Big Banged its way into our realities decades ago, but I know I am not alone. Many folks are regrettably still blissfully unaware of some of the big things taking place and shaping the computer-verse that will have life-changing ramifications for them, their children and all future generations.
Thankfully there’s a world of people familiar with what’s going on who don’t need to sign any corporate NDA’s to invent marvelous new computer-based toys and tools that will impact on our daily lives. Many are going to be great things that lead to even greater things.
Their use of AI, which many see as a “Super Hero,” will help save the world. The answers to nearly all of Earth’s issues can, in all likelihood, be solved by X number of generations of AI. It’s not fantasy. It’s also not guaranteed, but odds are in our favor that it’s true. Daily, I read about people of all ages getting comfortable with the notion of AI and incorporating it into “garage built” products. In turn, they’re using those products to springboard their prosperity to phenomenal levels.
Here are some examples, drawn from wired.com stories. Tom Simonite’s article entitled DIY Tinkerers gives some insight as to what’s possible using AI and a lot of desire. For instance:
Self-taught coder Robbie Barrat is using AI to advance his art. He is designing oddball fashions and having AI finish the lyrics of his latest original rap song, He said he trained his neural network on Kanye West songs. He started all of this in middle school. After picking up some coding knowledge, his AI could produce its own lyrics, often with vulgar lines, according to his teacher. After a stint with a self-driving car project in Silicon Valley, Barrat moved to Stanford University where he now works days in a bio lab. Barrat is still designing clothes with his AI and still writing songs with it.
In Oakland, CA, a fellow named Will Roscoe has come up with a vehicle he calls the Donkey Car.
A civil engineer by training, Roscoe was inspired to create the Donkey Car by a political defeat. Roscoe ran for a seat on the board of the Bay Area subway system, in 2016 and lost. Roscoe pledged as part of his campaign platform, to expand capacity or BART, by replacing trains with self-driving electric buses, but he finished in third place. Building his own pint-size autonomous vehicle seemed a good way to show voters that the technology wasn’t pure fancy
“I wanted to demonstrate it can work at a small scale,” he says.
As it turned out, his timing was perfect—a robotics hobbyist group dedicated to hacking RC cars was about to hold its first meeting in nearby Berkeley. There he met a fellow tinkerer, Adam Conway, who said he could help with vehicle construction. Roscoe, a self-taught coder, crafted its self-driving autopilot using TensorFlow, software created by Google and later released as open source. He also borrowed some neural network code from an attendee of the RC car meetup.
Roscoe’s final design learns to drive by watching a human steer the vehicle during demonstration runs. He named his creation Donkey Car after what he considers its spirit animal—safe for kids, not conventionally elegant and prone to fits of disobedience. But there is little to no doubt he will be successful at his autonomous vehicle project. He practically did it out of his garage and living room.
People are teaching AI to help in their dry cleaning business. One man began teaching AI to process, intake, catalog and give a receipt to the customer without a human being involved. It’s proven to be nearly flawless on all levels, so far. He intends to open a chain of automated dry cleaning establishments.
What was once considered impossible is now easily accessible to create mind-blowing programs. If you can imagine it, you can probably accomplish it with the help of a few neural networks and some determination. So get to it before someone else has the same idea and makes the millions intended for you.