Machines Want Your Job

I’ve been planning for a while to write a series of posts on this blog about how advances in science and technology have the potential to make certain current jobs carried out by people unnecessary. This topic gets some thoughtful coverage in the media from time to time but usually it’s just short throw away stories on some funny looking robot (ahem…see below). But the bigger picture for people’s careers and the economiy are important to think about as well. That is because if you know technology is being developed which could make your current job disappear in 5 or 10 years you may have time to consider a new career. Or imagine you see on the horizon that 20 years from now we may not need human bus drivers in cities because of self-driving cars. Then you wouldn’t encourage your children or any young person to get into that job for the long term, even if the benefits are great, because it’s fundamentally unstable in the long run due to technology.

So as a public service, and to give me a theme to write about, I’m going to try to regularly post here to raise awareness about advances in science and technology that have the potential to do away with entire careers completely. Or to give it a pithy Twitter hashtag:  #MachinesWantYourJob.

What to do about it

Some of these trends are obvious to a lot of people, such as repetitive factory work jobs being replaced by robotics. But a lot of cases aren’t so obvious. Personally, I think people are underestimating how disruptive self-driving cars will be on many jobs once the technological, safety and regulatory kinks are worked out. This may take a long time, but it’s progressing faster than some expected. Also, there is a lot of scientific research and technology development that is not so widely covered or understood by the media, so people don’t realize that some jobs could be just doomed in the long run. Another example is grocery checkouts, the current clumsy self-checkout lines in supermarkets are only a first step.  It is perfectly feasible with existing technology to build a supermarket or big box store with no checkout lines at all by using RFID tagged merchandise, QR code printouts from scales and object recognition on digital cameras. No cashiers would be needed, just pillars near the exits to confirm your purchase and pay.  It’s seems to be just a matter of time before it’s cheap enough that some store will implement it and do away with those cashier jobs.

Just to be clear, the intention in pointing out these trends is not to necessarily stop them in order to save existing, 20th century style jobs. The intention is to raise awareness about what may be coming and encourage people to prepare themselves for the future, to retool, to consider new careers while working in old ones so that when the hammers falls, they are prepared. Because it is unlikely anything going to stop these changes, and if you want to know you certainly can’t stop them if you don’t see them coming.

A more positive and inspiring way to think about this future as an opportunity to have many careers over your life is summed up much more eloquently than I ever could by this comic on Saturday Morning Breakfast Cereal, take a look.

Ok, Show us the Funny Robots

Entry number one – I hope you weren’t banking on being a noodle cutting chef in a Chinese restaurant, because the robots are all over that:

The best thing about this is how he felt the need to make it look like a 1970s stereotype of a robot. It’s actually really simple, barely a robot at all, anymore than the windshield wipers on your car are.

Now I’m hungry…noodles. mmmm.

See you next time, if you have any ideas for topics on this theme reply in the comments or tweet me @compthink.

Don’t Judge a Social Network by it’s Book Price

Far be it for me to defend Facebook…I think Google+ is a far superior tool for discussion and community and even Twitter is unmatched for certain types of interaction, as the Olympics showed clearly.

However, news of the impending death of Facebook because their share price is falling, are greatly exaggerated and miss the point.

Facebook does still have 1 billion users, and they’re active. Most of the people who use Facebook now are not early adopters or particularly savy with technology. They go to Facebook because all their friends and family are there. They aren’t going to leave even if something better comes along because there’s no reason to. You don’t go Facebook because it’s the Social Network you like best, you go to it because it’s Facebook.  So Facebook isn’t going to go away like GroupOn or MySpace anytime soon. (Even these supposedly  ‘dead’ sites  still have loads of users in fact.)

So this idea that since the share price of Facebook has dropped like a rock that the company is in trouble or that they are switching to ‘survival’ mode, seems ridiculous to me. Maybe Facebook isn’t a great investment right now, but that’s not because no one uses it. It’s a bad investment because it was insanely overvalued at the start.

NPR’s _Planet Money_ show had a great discussion about Facebook’s IPO the week after it happened (listen to it here: http://www.npr.org/blogs/money/2012/05/22/153300390/facebook-now-what).

They had one bit of analysis that went something like this, I’m paraphrasing: ‘In order for the initial IPO price of Facebook to be an accurate valuation of the company’s future profit Facebook would need to take over 90% of the advertising industry, not web advertising, ALL advertising worldwide including magazines, tv, newspapers, billboards.’

That’s simply ridiculous.

The initial price they were putting forward was ten times what you would expect from a reasonable web advertising play assuming you could squeeze as much money out of your stream as Google.  That means the people at Facebook thought they could do ten times better than Google at making users click on ads. It’s fine for them to believe that, go ahead, shoot for it, maybe Facebook can really invent a new concept of advertising that unlocks new, untapped money.

But the problem is  investors and pundits like the writer of this Globe article seem to think that initial price must have been a somewhat reasonable extension of current methods, but it wasn’t. So to their minds, if the price goes down by 50% from the initial price then it’s all over for the company. But 50% is still an overvaluation.

A decline in the stock price of Facebook to $10 or less does not mean they are dead. You know why? Because 1 billion people will still be using the site and clicking on some ads. And those people won’t want to switch to another network even if it is better. However, it does mean eventually Facebook is going to need to either find a way to make more money or accept a more modest chunk of the worldwide advertising pie.

Machine Morality

This post started as a response to a provacative question by David Brin on Google+:

It was in response to this fantastically titled article: “The Case Against Autonomous Killing Machines”

I don’t think we should be designing machines that kill people, period. But if you are going to have machine’s that kill sometimes and military uses fund a lot of this research they may not be willing to encode the Asimov laws. At least not in the same ordering (ie. following human orders will be more important than saving lives).

But a bigger question right now is whether an AI could reliably evaluate moral laws at all, even if it isn’t literally a killing machine. How sure does the system need to be that there is a human near by or that the machine’s actions will cause harm? How much failure are we willing to accept if it’s only 99% sure what it’s doing won’t cause harm?

Think of google’s self driving cars. That’s the nearest, potentially deadliest robotic systems we need to worry about in our daily lives. Any wrong move while driving could harm the passengers as well as people in other cars. You can never be _sure_ you won’t get into an accident. So how sure do you have to be? Then every once in a way someone will get killed anyways, who’s the blame? The robot? The engineer? No one? If we define these laws too strictly we may stop ourselves from creating amazing technologies that change the world. But if it’s too loose or not present at all then we enter this moral grey zone where accidents happen even though they might not have happened with more processing time.

Talking Robots and Psychedelic Drugs

Baby robot learns first words from human teacher – tech – 15 June 2012 – New Scientist

I’m always glad to see more methods from research in Artificial Intelligence/Machine Learning getting coverage in the media and being explained with some level of detail. Take a look at these two articles on applications of Artificial Intellgience methods in the study of the learning language in infants and in the effects of psychedelic drugs diagnosis. They give a nice high level overview of two powerful approaches that are not quite standard in AI and Machine Learning. The language learning robot is doing supervised learning with reinforcement learning approach where the agent randomly explores a landscape and weights good experiences to improve it’s model. The drugs study is applying a classifier to text descriptions about psychedelic trips and trying to predict the drug that causes it.

It’s that time of year again…password changing season.

If you use the popular professional social network LinkedIn you should go there right now and change your password to a new one, apparently some russian hackers got a few million passwords off their systems. Even if it’s a false alarm it’s a good idea to change your passwords every once in a while.

I would suggest coming up with a new password entirely. Why? because if you are like a lot of people you reuse a few passwords in many places, maybe even just one password. So the next thing you’ll want to do is go to every other important website,bank,computer you use with the same password as LinkedIn and change them too.

Having trouble coming up with a new password, here’s some good advice from XKCD:

If you want to know if your password was taken you can also try leakedin.org to check your password, you may want to compute your password hash offline though rather trusting their site. If you have python installed (if you have a mac or linux machine this should be true by default) you can run the following on the command line to get your password encrypted and enter the resulting string into the LeakedIn website. Replace “password” with your password:

python -c 'import hashlib; print hashlib.sha1("password").hexdigest()'

(Thanks to Nathan Taylor for that snippet of code)