More than meets the AI: Solving driverless cars (hint: it’s an infrastructure thing)

Here in my car

I feel safest of all

I can lock all my doors

It’s the only way to live in cars

In this issue:

  • A million AI robotaxis by 2020 (167 words)
  • Exit big tech (214 words)
  • Bonus: How to fix hyperloop (82 words)

A million AI robotaxis by 2020

Here in my car

I can only receive

I can listen to you

It keeps me stable for days in cars

Elon Musk once predicted there would be a million fully-autonomous robotaxis on the road by the end of 2020. Three years later and, as I predicted back when he made his ridiculous assertion, there still isn’t a single driverless vehicle in production. 

It turns out that the problem of autonomous driving was far more challenging than most of the experts had anticipated. Actually, to be more precise, teaching an AI to drive a car is a simple matter that was solved years ago. It’s teaching it not to kill its occupants, the occupants of any other vehicles, or any pedestrians that’s proven to be the big challenge.

The simple fact of the matter is that the majority of the major players who aren’t named Elon Musk have figured out how to solve the problem. But, because of what the solution would entail, they’re no longer interested. 

Exit big tech

How to solve the aforementioned driverless car problem? That’s easy. It’s in this issue’s title: build infrastructure for autonomous vehicles. 

This involves placing sensors in streets, walkways, thoroughfares, buildings, and other vehicles. We’d also need to ensure humans were detectable via sensors to vehicles, which would of course pose a significant number of privacy concerns.

But where it all breaks down is in the realization that big tech wouldn’t be able to monetize access to the technology required because such infrastructure would be considered a public service. This is why you don’t drink Google tap water or use Microsoft electricity services.

Don’t get me wrong, if there was a way to make it work, you would. 

But, if big tech can’t own the access then you can’t be the product. So don’t expect the same people working on driverless cars to have any interest in developing an infrastructure to support those vehicles. 

And that means you’ll be waiting around for Ford and BMW to finish developing their own systems. Or you could always just buy a Tesla. According to Musk, they’ll be able to drive themselves by 2015. No, that’s not a typo

Here in my car

Where the image breaks down

Will you visit me, please

If I open my door in cars?

Bonus: How to fix hyperloop

Real quick, this’ll just take a sec. While I was writing the above two sections I figured how to solve the hyperloop problem: fill in the holes with dirt, refund the cities that paid for this nonsense every penny that’s been swindled from them, and get a legal order directing said cities’ transportation managers to reinvest those funds into public transportation. 

You’re welcome. 

Here in my car

I know I’ve started to think

About leaving tonight

Although nothing seems right in cars


Read more More Than Meets the AI here!


We would love to hear your thoughts!

5 reasons why you should start a mini water propagation garden today