a thoughtful web.
Good ideas and conversation. No ads, no tracking.   Login or Take a Tour!
comment by veen
veen  ·  3174 days ago  ·  link  ·    ·  parent  ·  post: Google Steps Away from Humanoid Robot PR Problem

    Google is dead.

No, I think they made the right choice to sell out Boston Dynamics. BD makes robots for the military that can walk and can slowly do slightly more human things like interacting with objects. They're all-purpose robots, similar to the robots we all know from fiction.

Which makes sense when you are developing for the military, who hopes to reduce the workload of tons of people in tons of circumstances. They need robots that can walk on any terrain. However, the steps that need to be taken from the current robots to fully capable humanoid robots is enormous. We need leaps in both hardware (what BD does well) and software (what Google does well) to get there.

It won't translate well into a product in the next decade or two. Between then and now, I think robotics is much better suited for single-purpose robots that are more advanced, something like this or simple walking robots that can do very simple tasks.

My best guess is that selling Boston Dynamic is either

- Google figuring out that they don't really have a lot to contribute on the hardware side, or

- Google realizing that humanoid all-purpose robots are too far away from reality to be building them.

Also, if you think Google discontinues robotics projects because they might be a PR disaster, you haven't followed the discussions around their self-driving car enough.





bioemerl  ·  3174 days ago  ·  link  ·  

Not funding because it doesn't translate into a product is exactly why google is dead.

veen  ·  3174 days ago  ·  link  ·  

What did you think Google, a company, was doing before? Their moonshot projects are all products, they're just products that are an enormous financial risk to make. They're not a research group with infinite funding, no matter how much we want them to be that.

bioemerl  ·  3173 days ago  ·  link  ·  

    Their moonshot projects are all products, they're just products that are an enormous financial risk to make.

And it is likely that it is near impossible to know if a product will become a moonshot without knowing the future. Once you start killing projects because they "wont make profit" you stop being a company that pushes the border, and become one that is happy to remain inside of those borders.

veen  ·  3173 days ago  ·  link  ·  

    Once you start killing projects because they "wont make profit" you stop being a company that pushes the border, and become one, and become one that is happy to remain inside of those borders

...and one that is going to make sense. I'm not saying they should kill everything that doesn't go in the black soon, but I am saying that Google is still a company, not a charity. They didn't kill their AI department - only the part that makes specific humanoid robots. Google is much better at doing the software side, letting someone else figure out the hardware and logistical challenges. Case in point: Google car, Android, every Google app ever.

I see this more as Google focusing on where they can innovate the most. Which at the end of the day means they can push more borders and get more done, just not in the humanoid hardware business but in other things like AI.