Has Being Human Fallen Out of Style? Valerie Tiberius on AI and Friendship

The importance of friendship can be viewed in many ways, from its vital role in helping us survive evolution to a knowing look or a shoulder to cry on. What stays constant, however, is the human element. And yet, we now see this fact of life challenged by the rise of chatbots and artificial friendships, which points us to the question: What’s being human got to do with it, anyway?

Valerie Tiberius, a philosopher who studies well-being, has some ideas. She breaks three of them down for us here in her own words.

A good friendship

The best kind of friendship is good for its own sake.

I define an ideal friendship as an enjoyable, close relationship built on shared activities between people who care about each other for their own sake. Mutual care (concern, regard, or love) is crucial for this kind of relationship, which is something people value for its own sake, not just because of what it gets us. Companion chatbots do not — cannot — care about us (at this point in their technological development, at least), so companion chatbots cannot be ideal friends.

Good versus ideal

Friendship doesn’t have to be ideal to be good for something.

There are lots of human friendships that don’t have all the features of an ideal friendship. These relationships can still be good for us because they are useful or enjoyable, or because they help us learn things about ourselves and become better people. Companion chatbots do sometimes have this kind of value: they can be fun to engage with, they can give us helpful advice, and they help some people learn how to be better conversationalists with their human friends. So, while chatbot companions are not inherently good (not good for their own sake), they also aren’t necessarily all bad.

It gets messy

Even if friendships with AI are not all bad, there are many risks associated with them.

I feel deeply that a world in which companion chatbots replace human relationships is a dystopia, no matter how fun the chatbots are. Because the bots are so agreeable and flattering, people may be tempted to spend more time with them (and less time with disagreeable humans) than is good for us. 

This risk is particularly worrisome when it comes to young people who are learning what relationships are like and how to navigate them. Interacting with a sycophantic chatbot who never causes you any grief and who you can just turn off whenever you want, with no repercussions, does not prepare you for friendship with people. There are other risks too: corporate control, data privacy, chatbot psychosis, and disastrous advice that leads people to do harm to themselves or others.

Given all these risks, we need to do all we can to make sure the value of human friendship is recognized and supported. A good place to start would be not to take our own friendships for granted and to put some effort into being a better friend to other humans.

This story was edited by Deborah Sventek, an undergraduate student in CLA.