The topic of AI is not going to leave the zeitgeist anytime soon, and honestly, I don’t think it should. It is an important and urgent matter. But it is also somewhat confusing to talk about. The questions of AI span widely different discussions, and it’s easy to jump around between them and end up derailed. So, I thought it might be useful to break it down a little. First, let’s talk about some of the categories of conversations involved in this overarching subject, then look at some of the additional ways that those conversations can get confused.
The Capabilities of AI: One of the reasons this category of conversations is complicated is that the capabilities of AI are all over the map. The free version of ChatGPT might be making really obvious mistakes at the same time that there are headlines about rogue AI agents blackmailing CEOs. Additionally, AI has the simultaneous capacity for work that is profoundly good and work that is profoundly dangerous and destructive. So, this category includes but is not limited to topics such as:
- What are the benefits that AI might bring in the future?
- What are the helpful, humanizing ways that I can use it in my life right now?
- What destructive uses might AI be put to?
- What seemingly innocuous ways of using AI might end up undermining the kind of life I want to live?
The Power Offered by AI: This might sound the same as the previous one, but it’s important to distinguish them. The magnitude of AI capabilities, for good or for evil, will also have an impact on the power dynamics of the people who use it. That applies to us in our daily lives and relationships, to CEOs who are beginning to consider whether they can afford to pay a human for what a computer could do for them for free. When social media became our town square, there were suddenly private citizens who owned our whole means of public discourse. We have to expect even more dramatic power shifts and unintended consequences in a world woven through with AI. So, this category of discussions might include:
- Do these new thresholds of power require new legal guardrails? If so, what? And what would we need to do to implement them?
- What is job loss going to look like?
- Where do we want existing jobs to be protected and where is it ok for automation to take over?
- How do we minimize the negative ripple effects?
- What are the kinds of unintended consequences that this much power might unleash on the world?
- How can we make sure we are headed toward a future that prioritizes humanity?
What Does it Mean To Be Human and Why Does it Matter? The more sophisticated AI gets, the more sophisticated a facsimile it can create of human things. Coming amid an epidemic of loneliness, that makes the philosophical questions, “What does it mean to be human?” and “Why does it matter?” both urgent and practical. This is especially true of the friendships, romantic relationships and therapist relationships that real people already consider themselves to have with AI chatbots. This is also true in the realm of creativity. Films and writing, visual art and music have all been made by AI, and as the technology ramps up, the “creative” output gets more and more appealing. So, the questions of this category have wide-ranging applications, including but not limited to:
- What do we say to people who are desperately lonely and find comfort in the affirmations of a chatbot?
- How do we put up personal guardrails to keep our brains safe from the subtle dangers of a sycophantic digital companion?
- Which parts of our lives are we ok with digitizing, and which parts need to remain human?
- Is the value of art just in its aesthetic appeal or is there actually some importance to art being a place where you encounter another human being?
- Can we use AI to bring people together in real life instead of driving them further apart?
As if all of this was not enough to deal with, there are plenty of other factors to make the conversations even more convoluted. For example:
Distinguishing Between Global and Personal Stakes. Both of these considerations are important but in different ways. It’s good to be informed about the context we are in because it affects our lives and the people around us. For most of us, actually changing the global situation is way above our paygrade. But that shouldn’t make us think that we have no agency in how AI impacts our lives.
Philosophical Red Herrings. It is interesting to talk about questions of consciousness and whether or not robots would be malicious, but those questions do not address the more urgent questions of what immensely powerful robots can do, whether or not they have consciousness.
The Role of Fear. As Catholics, we put our hope in the Lord, and we don’t need to be afraid because he is already victorious. Conversations about AI have a uniquely dystopian flavor, one that we aren’t used to dealing with in our real lives. There’s a natural tendency to give in to how overwhelming it is by just ignoring the problem or assuming it must be overblown. But courage doesn’t mean ignoring risks. Fear, rightly ordered, gives us important information about risks so that we can proceed with prudence. The virtue of courage lies between two extremes: on the one hand, allowing fear to lead and, on the other, ignoring fear entirely. What courage looks like in the realm of AI isn’t simple or obvious, but it’s important to seek that middle ground so that we can be both brave and wise in how we deal with this technology.
