On Thursday, at an event called "A.I. Day," Tesla's CEO, Elon Musk, doubled down on the company's artificial intelligence (A.I.) efforts. The most interesting thing he announced had nothing to do with Tesla's cars, however. Musk says the company is planning to build an A.I.-powered humanoid robot.

I think it's important to be clear that Tesla's livestream wasn't really a product launch, as much as it was a recruiting event. In that sense, it was brilliant. Building a self-driving car isn't nearly as exciting as working on a robot that lives in your home and can do things humans can do like clean, cook, or go to the store for eggs. 

 inline image

Certainly, if you're trying to improve your A.I.-computing capabilities, getting people excited about edge cases is a good way to recruit smart people to your cause. And, if the problems Tesla has had hitting its own goals is any indication, the company could use all the help it can get.

For example, Tesla is facing an investigation into its autopilot feature over a series of incidents where the company's vehicles collided with emergency vehicles with their lights flashing along the side of the road. Will a Tesla Bot randomly run into your couch or table? 

Musk also promised there would be a million fully autonomous vehicles on the road by the middle of 2020. That still hasn't happened. Eventually, all of those broken promises add up and start to damage your credibility. That's a problem for obvious reasons. Finding smart people to help Tesla solve the challenges of A.I. is a smart move--especially if it means that it can actually deliver on those promises.

There is another problem, however. The fact that Musk had to clarify that you'll be able to outrun the bot, and that you should have no trouble overpowering it, says a lot about the level of anxiety most people have about inviting an artificially intelligent computer shaped like a human into their home. As cool as it might sound, I'm not sure the world is ready for that. 

I'm also skeptical that--assuming Tesla ever delivers this product--it will lead to the A.I.-driven utopia that Musk is promising. Musk seems to think that having a robot that can handle trivial labor will lead to a future where "physical work will be a choice." Of course, in this version, everyone would be supported by a universal basic income from the government. There are a lot of dots to connect there, and almost none of them seem likely anytime soon. 

What does seem likely would be that instead of a robot that can take care of mundane tasks like grocery shopping, corporations and ultra-wealthy individuals would simply buy them all to replace workers on assembly lines, in grocery stores, or at restaurants, putting millions of people out of work. That's not exactly utopia.

That's why I have a hard time deciding whether the entire thing is terrifying or brilliant. I suppose it's both. For Tesla, it's brilliant if your goal is to attract attention and the talent you need to make it happen.

For the rest of us, I'm comforted by the fact that the chances of seeing one of these walking down the aisle in the grocery store toward me are practically zero. That's a good thing because the thought of that is absolutely terrifying.