The Importance of Computer Ethics

The word ethics is coming up more and more often these days, driven by, among other things, the questions raised by artificial intelligence. This may seem essential, but it is a subject that all of us as computer scientists should have tackled much earlier.

Ethics — a Definition

If we had to settle for a simple definition, we could say that ethics is a philosophical discipline dealing with judgments. Ethics, therefore, differ from morality in the sense that morality is relatively defined, and potentially specific to each individual, whereas ethics consists precisely in studying and reflection to learn what is moral or not. The ethical approach can diverge even more widely between individuals than morality, in the sense that it is a question of defining not what is right or wrong, but of defining the intellectual framework within which we define what is right or wrong. This distinction is very important to make, in the sense that these two words are often confused with each other.

The Ethics of Computer Science

It is indeed a topical issue, and more and more conferences and think tanks are turning to it. However, it is not a totally virgin subject: books such as The Hacker Ethic or The Cathedral and the Bazaar have already approached this subject. Unfortunately, these books do not focus on the end use of computers, but rather on computer scientists' motivations and the associated organization of their work.

What We Lack

So how do we know which technology is ethical and which is not? As far as I am concerned, I consider that it is the use of the technique, and not the technique itself, that is either ethical or unethical. An anecdote will enhance this point of view. A few years ago, I attended a presentation by an editor of a Complex Event Processing (CEP) solution, a solution often used for fraud detection. My client at the time being an insurance company, there was a good chance that it could be interesting, and concerning my own morality, I found this kind of computer solution to be particularly moral. But during this presentation, the publisher told us how proud he was of the success of a project to detect fraud involving reductions on the back of receipts, by the cashiers themselves. Poor people with very low wages, who were clearly seeking to improve their condition. Even though it was illegal, the idea of firing the cashiers concerned, who just wanted to get 15 extra euros per month, seemed abject to me.

I bring this up, not to share my moral point of view, which concerns only me, but to share the effect of the surprise that there may be to discover the use that can be made of technologies that may seem ethical, and that by their use become the opposite. I was not prepared for this as a surprise, because no one had prepared me for this, including myself. Indeed, higher education often lacks any approach to ethics and morality, and it is up to us to make our own reflections. And it quite unfortunate if our work is eating away at us from within because of its negative impact on society. Just ask Alfred Nobel.

The Urgency

One might argue that this problem is not a new one, and that if we have survived it so far, there is no hurry. I would argue that we should, as a society, account for the negative effects of what has already been done, but it may be a lost cause. On the other hand, we have the subject of artificial intelligence, which opens up a very wide range of applications that can be frightening. Remember that video of that scientist calling for a moratorium on killer robots? It's a bit scary, but we could also open this subject up to possible future police robots. What "morality" should be implemented in these robots? In fact, all technological advances bring their share of ethical concerns, as in the case of autonomous cars. In short, each advance on artificial intelligence brings its own set of questions.

It is, therefore urgent, that computer scientists of all kinds appropriate the subject of ethics and morality; we have fallen far too far behind, and we are certainly at the forefront of the technology that breeds these conversations.