Google self-driving car pioneer wants to teach people how to face the future • Hacker News Board

Re: Google self-driving car pioneer wants to teach people how to face the future

#41

Post by sandworm101 » Fri, Sep 04, 2015, 5:26 PM UTC

davidcgl wrote: post #35

sandworm101 wrote: post #22

Arrogant, as is typical of those with a financial stake in a particular version of progress. This guy is under the belief that his version of the future will happen and that it is his job to help others realise that truth. Technological progress may be inevitable, but that progress envisioned by any particular person is not. Nanodegrees may seem a good idea, and they no doubt make sense financially to google, but tha…

Nanodegrees may seem a good idea, and they no doubt make sense financially to google Google doesn't own Udacity.

Yes but Google does higher lots of young people. They would rather higher kids not saddled by debt. Recent grads without debts will settle for lower salaries than those who need to pay down massive student loans.

As a general rule, industry prefers narrowly-targeted education for skilled/unskilled workers. But when it comes to managers those four/six/eight-year degrees from centuries-old universities are all important. You don't see many executives without degrees. Founders may forgo education, but not so much the people they higher to run things.

Re: Google self-driving car pioneer wants to teach people how to face the future

#42

Post by StavrosK » Fri, Sep 04, 2015, 5:30 PM UTC

JoshTriplett wrote: post #38

StavrosK wrote: post #36

Earlier quoted context omitted.

> But at the same time, no job will need to Why? Because this AI, which is basically just people smarter than us and getting smarter by the day, will decide that its only purpose in life is to serve its idiot masters?

Because if we build it correctly, it won't be a person at all, and won't have a value function orthogonal to our own. Building something that can self-improve and solve problems doesn't mean we need to build something that has independent will, motivation, or values. And it's a lot safer not to.

Sure, but "an AI that's smarter than humans" implies that. I'm sure we'll build many algorithms that are better than us at specific tasks, and we already have.

Re: Google self-driving car pioneer wants to teach people how to face the future

#43

Post by Florin_Andrei » Fri, Sep 04, 2015, 5:30 PM UTC

bsaunder wrote: post #2

I agree with the premise that machines will increasing exceed our capabilities (and are improving quicker than we are). This seems so obvious as to be not debatable. I also think there are a lot of merits to online-education and Udacity seems to have the right general idea. On-line education needs to be more engaging and entertaining than simply videos and online collaboration. However, this whole "people need more s…

> In fact this push for increased human skills seems like it will only widen the gap between the uber skilled and the 99%.

What happens is that pressure is simply increasing all the time, and people are on a bell curve w.r.t. how well they can keep up. We are effectively grabbing that bell curve and pulling its ends much further apart.

We need a solution, otherwise things are going to get very, very ugly.

Re: Google self-driving car pioneer wants to teach people how to face the future

#44

Post by Florin_Andrei » Fri, Sep 04, 2015, 5:32 PM UTC

hiddencost wrote: post #9

mathattack wrote: post #5

Earlier quoted context omitted.

Many great scientists (Einstein amongst them) get into trouble when they veer into Economics. Einstein predicted social unrest from massive unemployment due to productivity improvements. Productivity improvements help people in aggregate. The question here is will a small subset of the world accrue all the benefit from the AI productivity improvement? One answer is basic income, presumably based on taxation of the ri…

Russia tried that. When the USSR ended, they gave every one vouchers to bid for shares of state owned companies. The little people had no idea what they were doing and got screwed brutally. It created many of the ridiculously wealthy Russian oligarchs you see today. https://en.m.wikipedia.org/wiki/Privatization_in_Russia

It could be argued that the population over there was very unprepared to make these choices - 3 generations of communism tend to shape your thinking in interesting ways.

Re: Google self-driving car pioneer wants to teach people how to face the future

#45

Post by arstin » Fri, Sep 04, 2015, 5:37 PM UTC

If Thrun is right about AI "outsmarting people in every dimension", how in the world could more entertaining job training videos organized into "nanodegrees" help us at all? I know this article was just a free ad for Udacity so he had to say self-serving things, but do you think the dude really believes this???

As others have pointed out, the real problem before us this century is to somehow decouple having a basic standard of living from performing work---more specifically from "contributing to productivity". My amateur guess is that even with our current state of food and energy production, the biggest barrier isn't the massively difficult economic and organizational problem, but advancing as a culture beyond our entrenched moral assumption that the means for basic living is something to be earned rather than a human right.

Re: Google self-driving car pioneer wants to teach people how to face the future

#46

Post by filoeleven » Fri, Sep 04, 2015, 5:39 PM UTC

Mimu wrote: post #3

bsaunder wrote: post #2

I agree with the premise that machines will increasing exceed our capabilities (and are improving quicker than we are). This seems so obvious as to be not debatable. I also think there are a lot of merits to online-education and Udacity seems to have the right general idea. On-line education needs to be more engaging and entertaining than simply videos and online collaboration. However, this whole "people need more s…

I agree with that. Not sure if the basic income is THE solution, I didn't think about it too much, but I agree with everything else. People don't need jobs, they need money. Jobs are the only legal way to get money but money is the issue, not jobs.

> People don't need jobs, they need money. Jobs are the only legal way to get money but money is the issue, not jobs.

People don't need money, they need a level of security in the basic necessities: food, water, shelter. Money is the primary legal way to get these things, but the basic necessities are the issue, not money.

Basic income, or any good solution involving money, is pragmatic and worthy of consideration since money is so embedded in the culture, but when we are talking about real needs we're talking about physical needs, of which money is not one.

Re: Google self-driving car pioneer wants to teach people how to face the future

#47

Post by criddell » Fri, Sep 04, 2015, 5:54 PM UTC

JoshTriplett wrote: post #32

criddell wrote: post #6

Earlier quoted context omitted.

> We need to pivot our economic system towards a basic income model. So putting a floor on what people get each year seems like a pretty good idea. How about a ceiling as well? What's an equitable spread? 10x? 100x? Should the basic income given to residents of Manhattan be 2.5x what people in Wichita get?

> How about a ceiling as well? What's an equitable spread? In no possible way up to you to decide, thankfully. It's one thing to say "nobody should go hungry or live on the street"; we could argue about the most efficient way to go about achieving that, but it's not an unreasonable statement. It's quite another thing to say "you're not allowed to be too successful or we'll steal it from you".

It isn't up to me to decide, but as a participant in our democracy, I have a say just like you.

> you're not allowed to be too successful or we'll steal it from you

Well, maybe there wouldn't be a hard ceiling, but I could see the revival of the 90% tax bracket for big earners.

Re: Google self-driving car pioneer wants to teach people how to face the future

#48

Post by criddell » Fri, Sep 04, 2015, 5:55 PM UTC

yazaddaruvala wrote: post #39

criddell wrote: post #6

Earlier quoted context omitted.

> We need to pivot our economic system towards a basic income model. So putting a floor on what people get each year seems like a pretty good idea. How about a ceiling as well? What's an equitable spread? 10x? 100x? Should the basic income given to residents of Manhattan be 2.5x what people in Wichita get?

Governments shouldn't provide money. The ideas behind basic income are to ensure the majority of people are one step higher on the hierarchy of needs. That said, it would be best for governments to provide free safe housing, heat, sanitation, food, water, communications (fast internet and telephone), medical care (including preventive care through gym memberships and such), and high quality accessible education, cert…

> That said, it would be best for governments to provide free safe housing, heat, sanitation, food, water, communications (fast internet and telephone), medical care (including preventive care through gym memberships and such), and high quality accessible education, certification.

That makes a lot of sense.

Re: Google self-driving car pioneer wants to teach people how to face the future

#49

Post by danans » Fri, Sep 04, 2015, 5:56 PM UTC

criddell wrote: post #6

bsaunder wrote: post #2

I agree with the premise that machines will increasing exceed our capabilities (and are improving quicker than we are). This seems so obvious as to be not debatable. I also think there are a lot of merits to online-education and Udacity seems to have the right general idea. On-line education needs to be more engaging and entertaining than simply videos and online collaboration. However, this whole "people need more s…

> We need to pivot our economic system towards a basic income model. So putting a floor on what people get each year seems like a pretty good idea. How about a ceiling as well? What's an equitable spread? 10x? 100x? Should the basic income given to residents of Manhattan be 2.5x what people in Wichita get?

Basic income isn't intended to pay for luxuries such rent/real-estate in Manhattan. If one wants to live a very high quality of life in a desirable place, one will still have find a way to make enough money to afford that outside of basic income.

The idea is to lift the very bottom of the income distribution, those who are likely to be put out of work first by automation, to a level where they are not plagued by a desperation for basics that causes them to damage themselves and society (i.e. drug abuse, crime). People in that situation already represent a significant cost, both in dealing with the problems, and also in the lost human potential.

The hope is that as automation reduces the cost of a lot of basic consumables (not rent or real estate), fewer people are kept in the aforementioned desperation.

With automation, these consumables will also not vary tremendously in price between Manhattan and Wichita, since the one of the biggest differences in production costs between both places, labor, has been removed from the production costs. Therefore, there would be no need to index BI by location.

Basic income can't solve the problem where automation has effectively removed all human-added value in all jobs. At that point, we'd have to impose more dramatic measures of redistribution, and also come up with new ways of measuring individuals' contributions to the general welfare, and reward them based on that. But I think we're still far away from that.

Re: Google self-driving car pioneer wants to teach people how to face the future

#50

Post by analyst74 » Fri, Sep 04, 2015, 6:13 PM UTC

mathattack wrote: post #5

bsaunder wrote: post #2

I agree with the premise that machines will increasing exceed our capabilities (and are improving quicker than we are). This seems so obvious as to be not debatable. I also think there are a lot of merits to online-education and Udacity seems to have the right general idea. On-line education needs to be more engaging and entertaining than simply videos and online collaboration. However, this whole "people need more s…

Many great scientists (Einstein amongst them) get into trouble when they veer into Economics. Einstein predicted social unrest from massive unemployment due to productivity improvements. Productivity improvements help people in aggregate. The question here is will a small subset of the world accrue all the benefit from the AI productivity improvement? One answer is basic income, presumably based on taxation of the ri…

When people think state owned enterprises, they think the failed USSR and communism ideal.

A more recent and viable model to look at is probably some of the successful Sovereign wealth fund, like Norway, UAE, Qatar and etc.