Bob Dylan sang in the sixties to articulate the then impending social change: " People get ready, there's a train a-coming, you do not need no baggage, you just get on board. " The artificial-intelligence Train also left and put course to the promised land. All we have to do is to step aboard. Artificial intelligence allows us to realize our dreams: it has for instance to cure potential diseases, eliminate world poverty and climate change. The promised land may sound tempting, but so far the general public is still reluctant to get on board. In an article in Wired gives President Barack Obama his views on a number of issues around AI and what restrains people. In this blog I am going deeper into some of the ethical issues that appoints President Obama.
Fear of AI is more than an "image problem"
Despite the enormous potential of AI to solve big problems, many people are afraid of. This fear is the past decades, partly fueled by the many films in which super intelligent robots turn against humanity and want to take over the domination of our planet.
But it's not just films that warn us against the dangers of AI. It's also a number of very prominent people, such as Bill Gates , Elon Musk and Steven Hawkins . This indicates that some reason for concern it is in place.
The tech giants, some of which have invested in recent years in many AI, recognize the problem and have taken steps to do something about it. Amazon, Facebook, Google, IBM and Microsoft have joined forces and founded a new organization, The Partnership on Artificial Intelligence to Benefit People and Society .
One of the objectives of this organization as much as possible to inform people about AI and to provide a platform for public debate, especially about the dangers and ethical aspects.
Artificial general intelligence (AGI)
In the interview, President Obama makes the distinction between " narrow artificial intelligence (AI narrow) and" artificial intelligence generalized "(AGI). With AGI's artificial intelligence is meant as we know from the movies. Eg Skynet , the computer thinks for itself from the Terminator movies, which is against humanity.
Among the experts is currently the consensus that we are in the short term no worries about having to make this form of AI. There is still a number of technical breakthroughs to build AGI.
I am addressing in this blog was not the ethical issues surrounding the future Skynet, but to those around " narrow AI .
Narrow artificial intelligence
While AGI is still something for the future, narrow AI is already applied. It contains no general intelligence, but uses algorithms and computers to perform complex tasks such as face recognition, driving or finding information on the Internet.
If properly harnessed, it can generate enormous prosperity and opportunity. But it also has some downsides That we're gonna have to figure out in terms of not Eliminating jobs.
The danger of exclusion
AI takes its analyzes a large number of variables, and calculates the most obvious choice. This can lead people to a standard profile are less likely, for example, to find a job or to take out a loan.
Axis technologies emerge and mature, then figuring out how They get incorporated into Existing regulatory structures Becomes a Tougher problem, and the overnment needs to Be Involved a little bit more.
In an "inclusive" society that does not want people to keep out, this is not desirable. In the interview, Obama indicates that the exceptions are precisely that make us human. The weirdos, freaks and wierdo's developing a brilliant invention or theory against existing conventions and thereby help mankind further step - think of Albert Einstein, Charles Darwin or Mozart.
Perhaps there is a task for our government is to ensure that the use of AI not exclude people because of their job opportunities or by reducing access to credit.
AI can not be held responsible
An additional issue is that of responsibility. It is annoying when you are rejected for your dream job, it's even worse when you find out that this has been done by one person but by an artificial intelligence bot . If you then want to know what the reason was for the rejection, is the second problem: " Nobody knows ."
AI systems would act as a black box . You can see what the output is, but you can not look inside the black box itself and trace how the system has come to the decision. In the spring of this year, the European Union has adopted a law that gives its citizens the right to an explanation of how an institution has reached its decision, even if that decision came about by an AI.
Such a statement is also important for the acceptance of AI by the general public. Google is currently investigating whether the images from her " deep dreaming 'program can be used to indicate errors in AI and thus gain insight into the workings of the black box.
AI can we leave decisions about life and death
At this time driving all self-propelled cars Google around and Uber does all tests with self-propelled taxis in Pittsburgh.
But Joi made a very elegant point, All which is, what are the values That we're going to embed in the cars?
To introduce large-scale self-propelled cars also poses a number of ethical issues. How the self-propelled car will, including a family react when suddenly another family cross the road? What choices does it in situations where it must choose between life and death? How to submit the accounts for the choices made?
increasing unemployment
The use of AI will make processes more efficient at lower cost. Take the example of Uber and the self-propelled taxis: using AI taxi industry will be many times more efficiently, at significantly lower cost. The downside is that many drivers will lose their jobs. AI eventually will cause productivity and increase prosperity in our society, but it decreases employment.
To most of the 20th century led automation to increase productivity, which in turn led to new jobs. At the end of the last century there emerged a fracture in that trend. Productivity is still increasing, but the job anymore.
A rising prosperity combined with fewer jobs requires a different structure of our society. We leave the era that wealth and income are distributed through labor. Therefore this will increasingly take place through ownership (equity). We will have to find a balance between work and distributing the increasing prosperity.
What is indisputable, though, is That axis AI gets Further Incorporated, and the society Potentially gets wealthier, the link between production and distribution, how much you work and how much you make, gets Further and Further attenuated-the computers are doing a lot of the work.
Get on board
We are currently on the threshold of a new world in which AI will play an increasingly important role. To ensure that everyone gets on the train, is an open and fair information is of great importance. The US government and big technology companies recognize this and have already taken steps to inform as many people and involve them in the discussion.
In particular the provision of information and discussion about the dangers and ethical aspects of AI are important. Norms and values are the basis of the decisions we make. Crucial for acceptance is that decisions reflect an AI system takes the values of our society.
This information and discussion I miss in the Netherlands. I hope I have given a better information regarding this topic and also have spurred an impetus to write this blog for more discussion.
No comments:
Post a Comment