Today I want to discuss the quality of life in the age of AI. To frame this conversation, let’s consider an extreme scenario that I don’t believe will happen, but that can help clarify our thoughts. Imagine a world where technology has advanced to the point where we have extraordinary AI and robots. People use these technologies to create robotic AI partners. Now, imagine you have a choice between a human partner and a robotic partner. The question is: Which one is better?
Let’s assume that the human partner will have ups and downs, strengths and weaknesses. In contrast, the robot may never reach the highs of a human but also won’t hit the lows. If humans range from amazing to terrible, robots will consistently be just good.
So, we have this idea of a robotic companion versus a human companion. The human companion has the potential to be exceptional, while the robotic companion can’t be terrible. They’ll always be nice and kind, and never complain about anything.
Now, imagine waking up in the morning with a choice: Whose house do you visit and who do you spend time with—the robotic partner or the human partner? My concern is that people might often choose the robotic one because it’s easier. “Do I really have the energy to deal with complaints today?” they might think. The robot never complains. Thus, people might opt for something consistently okay rather than risking something potentially bad but also potentially great.
It’s not just about one choice. What happens if you keep choosing the robot over time? After two weeks of visiting your robotic partner, how will your human partner react? Likely not kindly because you need to invest in relationships for them to be good.
One possible outcome is that we’ll take the easy escape route whenever relationships get challenging or demanding. We might avoid risks and get worse at maintaining meaningful human connections.
Will we actually have these robots? Maybe. Maybe not. Even if we don’t, the thought experiment is important. Every day we choose between engaging with people on Instagram or fostering real human relationships. Do we prioritize long-term benefits or short-term comforts? Are we opting for experiences that offer a range of outcomes, including incredible highs, or settling for something consistently mediocre but free from downsides?
Spending a night on Instagram isn’t terrible; it’s not amazing either. It’s just okay. My worry is that AI will offer us many moderately good options, and because we dislike negatives, we’ll often choose these safe options. In doing so, we might miss out on amazing experiences with other humans.
Will we end up more lonely or less lonely? I think we’ll find many, weak substitutes that help in the moment but leave us worse off in the long run.
Is this a future I’m looking forward to? I’m curious about how it will unfold, but I don’t think it would be good for humanity. Substitute relationships, whether with robots or just AI, aren’t a positive step in human development. They would provide an easy escape in the short term but would be destructive in the long term.
