I just finished The Fourth Age: Smart Robots, Conscious Computers, and the Future of Humanity, and it was one hell of a read.
A lot of us are concerned about the coming of robots that arguably will be substantially smarter and more capable than we are. This book likely will make some feel a tad more comfortable about the future, because it points out in some detail that a lot of the blue-collar jobs that supposedly would be going to these metallic false humans won't be.
On the other hand, a lot of the white color jobs we thought were safe -- well, they probably won't be around for long. Those of you who are in the fields where much of this coming effort is focused, like medicine -- well, you are either screwed or you're going to have an exciting time, depending on how much you like change.
I should point out that while there is a lot in this book that is impressively right, there also is a lot that is wrong. It showcases that the level of change we will be facing in the next decade or so will be so disruptive that no one, even the incredibly well-connected author of this book, Byron Reese, is going to be able to predict the future with much accuracy.
Yet this book lays a great foundation that will help you understand the changes that are coming, and understanding a change often goes a long way toward removing the fear surrounding it. It is survivable -- and I agree with Reese's point that there are stronger odds in favor of the future being far brighter than today than the other way around.
If widely read, books like this can become self-fulfilling prophecies. It can help ensure that bright future, which is one of the strongest reasons not only for reading it but also for getting your friends to read it.
I'll share some ideas on how artificial intelligence will impact the future of your career, and then close with my product of the week: a really handy International WiFi service that substantially could reduce your connectivity costs when you travel overseas.
AI and Your Job
It is really interesting to me how perceptions surrounding which jobs are most at risk have changed over the last five years. When I first started seriously covering AI, the consensus appeared to be that smart robots would take the jobs of blue collar workers first.
The logic was that those jobs were relatively easy to perform, the workers often were problematic because they really didn't enjoy their jobs, and, honestly, I think the folks making the predictions were biased in thinking the robots wouldn't take their jobs.
The book does the best job I've seen so far with regard to coming up with a way to calculate job risk -- and if you use the formula from the book, blue-collar employees are relatively safe. Since I want you to read the book, I won't share the entire formula, but in summary it suggests that the more expensive the job, the more knowledge it requires, and the more people who do it, the more likely it is that the job will be done by a future AI. This is economics at work.
AIs are great with anything that has to do with data, but they aren't good with breadth, particularly if there is a lot of random activity. The economics favor replacing highly paid workers at scale over low paid workers.
Looking the other way, the more variables you have, the closer to unique your skills are, and the less you make, the less likely it is that you will be replaced by an AI. For example, a radiologist is an ideal target for an AI while someone who repairs old clocks is not.
However, I should point out that the folks creating AIs often don't use this type of mechanism to decide on the economics of the result. For instance, another job that the book suggests should be at low risk is actor.
You might remember that a few decades back there were commercials featuring John Wayne that went up long after John had gone to that great western movie in the sky. Old footage of other deceased actors was repurposed in other such commercial efforts as well.
Now there is a substantial ongoing effort to create digital actors with their own performance capabilities.
Another profession that shouldn't be on the short list is truck drivers. The complexity of driving trucks in traffic is very high, truck drivers aren't that expensive, and while there are a lot of them, they have a powerful union that will resist the effort. When you factor in the cost of replacing the truck, the change is relatively expensive.
However, the U.S. is short something like 50K truck drivers, and folks have been predicting that if we don't fix this, it will drop the country into a recession. So, there has been a ton of effort expended to create autonomous trucks, because the need is so great. (Interestingly, this could be offset using immigrants -- particularly given that the trucking companies are willing to fund training and living costs during the training because they are so desperate -- but sadly, our government has been blind to that solution.)
This means that while the formula in the book can help you feel safe, you still need to maintain a high level of awareness of what is going on with AI research in your field, because the clear majority of folks building AIs clearly haven't been using this formula.