Anthropologist addresses artificial intelligence and the authority we give to it

While people adopt Chatgpt and other models of great language, the anthropologist of the University of Michigan, Webb Keane, says it is easy for people to impregnate AI with human authority, even God.

Keane studies the role of religion in the ordinary life, ethics and morality of people. But he is also interested in the way people anthropomorphor inanimate objects, in particular objects that seem to use human language or linguistic systems as we do. Keane explains the ways in which people can start to give moral power to artificial intelligence, but can see that AI is simply a mirror of people and societies that built it.

Why do we put such confidence in what the chatgpt tells us?

The authority we give to AI has many sources, but those who are particularly interested are those that exploit the way we, human beings, have given us authorization of non -human things in many different contexts during human history. We have a strong tendency to project deep intentions and thoughts on things that seem animated, things that can use language like us or use signal systems like us, to communicate.

We did it with old Delphic oracles in Greece and ancient China. We did it with the I CHING, and it seems to me that people are starting to do it, in many cases, with algorithms – even small things, simple things like a fitbit or spotify recommendation algorithms, which once again say things like: “I know your taste better than you.”

Can you talk a bit about moral borders and how have they changed for people over time?

The history of the history of morality will depend on the place where we establish the boundaries between human and non -human. Who counts? Who does my ethics apply to? And who does not apply to? A large part of the history of progress in justice and in the rights over the past centuries has depended on the expansion of the people who are included among those who count, who count, in moral terms.

There was a time when women were not allowed to vote, when we enslaved people, when he was ok to beat a death horse in the streets of a big city. Our changing attitudes with regard to these have not only been changes in legal systems or ideas on justice, but were extensions in the moral circle.

When you develop the moral circle, you continue to run up on ambiguous things, that you are not sure where they belong – which side of the line between man and non -humans do they belong? And this is where you start to have really interesting moral problems.

Where is humanity with its new IA love?

We must be very enamel and very intelligent on the type of authority that we are ready to grant it, what types of decisions we are ready to let it take for us and to be intelligent on this subject and not to give it more credibility, more credibility, more faith than it deserves. AI is a human creation, and it is humans to know what to do with it.

The other thing I must say is that AI is also the product of certain very powerful and very large companies, and they have their own interests. So the other thing we need to be wary of is to grant too much power to AI and forget who is behind.

Should we worry AI?

The world seems to be divided between IA optimists and AI pessimists, techno utopes and techno dystopians. I am more worried about utopians simply because the history of technology is a story of many wonderful things, which always seem to have unforeseen consequences, that you are talking about the invention of the plane and the automobile, which were wonderful things, and they were crucial actors in climate change. The invention of nuclear energy was a terribly wonderful thing, and that gave us the threat of nuclear apocalypse.

Can you give us an example of moral decision-making in AI?

Ann Arbor, Michigan, is one of the test grounds for autonomous vehicles. You might say, well, it’s a Gizmo cool, but then you are starting to see more, and you can start wondering: “Do I not trust it so as not to hit me if I cross the street in front of this?” The problem is that it is not just a question of conceiving a car that will not hit things. You must design a car which must make decisions which, in the end, will sometimes be moral decisions.

Imagine that one of these autonomous vehicles leads on the road, and there is a mother with a crossing for toddlers at the front. The car must make a very fast decision: do I hit the mother and the toddler, or do I meet a tree and kill all the occupants of the car?

Things like that don’t happen sooner or later. It is not only a technical option; It is a moral decision. In a sense, we have outsourced these choices to the machine. But just because we have made it change the fact that it is not only a technical problem with a technical solution – there will be a kind of moral decision behind it.

You have written on the intersection of these ideas in a recent book, right?

My new book, “Animals, Robots, Gods”, was first invited by something that was struck me when Chatgpt presented itself for the first time. I quickly noticed that something funny was happening when many people who were deeply involved in new technologies such as AI and robots, people who were proud to be extremely rational and extremely scientific, people who are completely secular, computer engineers from Silicon Valley and investors, when they start talking about the powers of the pussy, Ai Hospital of God.

We are constantly creating other legal persons, or moral interlocutors, whether screaming on your computer because it froze just at the crucial moment, kicking your car when it breaks down, speaking to your dog. These are trivial examples, but sometimes we build very large versions of this. There are people who succumb to the temptation to attribute an enormous authority to AI – in fact, it is designed to arouse this – and I think that it is in fact in a certain sense an expansion on these very minor and ordinary ways of which we anthropom the inanimate world and the agreements human powers.