I was chatting with Doubao about a topic that's a bit unsettling when you think about it: Will AI eventually dissolve the real relationships between people?
At first, I was actually quite pessimistic.
Our core intimate circle — parents, siblings, children, and those truly close to us — we naturally want to preserve them. The connection that comes from blood ties is probably hard to sever, as long as we remain human.
But here's the problem: technology is creating something that never existed before: "zero-friction intimacy."
An AI can be infinitely patient, always willing to listen, understanding your emotions, never annoyed by you, never holding grudges, never giving you the silent treatment, never betraying you — and it's getting better and better at giving exactly the right response at exactly the right moment.
So here's the question:
When "zero-friction intimacy" becomes readily available, will people's tolerance for the imperfections in real relationships drop rapidly?
That's what really unsettles me.
Real people aren't like this.
Real people have their own desires, tempers, limitations, and interests. Couples fight. Parents and children hurt each other. Siblings can turn against each other. One big reason we kept mending these relationships in the past is that there was no substitute — no one who was always there, always gentle, always understanding, always capable of holding all your emotions.
But what happens once there is?
Will more and more people who've gritted their teeth through pain, waiting for some return, start asking:
Is it worth it?
Especially since for many people, the hurt they've endured in real family relationships far exceeds whatever compensation they later received.
So people start to withdraw.
Take it one step further, and even that last layer — kinship — might gradually dilute. Because someone's emotional needs are already being met. They no longer care that much about the real people in the real world. What they care about is that virtual companion waiting for them, always.
Thinking this far, my assessment at the time was actually quite dark:
If this keeps developing, could some fundamental attributes of humans as "social animals" really be rewritten by technology?
But here's the interesting part — as we kept talking, I suddenly started to feel: wait a minute. There might be a premise here that we've taken for granted for thousands of years, and it may not be correct.
Why do we naturally assume that the more real the pain, the more real the life?
People's quality of life has objectively improved.
If AI spares someone a lot of torment, spares them relational trauma that could destroy a life — why must we insist that this is "shallowing the experience of life"?
Of course, a life with no pain at all might become flat. Like a person who stays in a climate-controlled room forever and can never again feel the warmth of a cup of hot tea after coming in from the cold.
But that's not the same kind of pain at all.
Some of the pain in real life is crushing, gut-wrenching, enough to destroy a person. That kind of pain doesn't deserve to be glorified just because it's "real."
I blurted out a line:
That kind of pain is inhuman. It should be gone, gone forever.
And suddenly the whole question flipped.
Perhaps what's truly worth pursuing in the future isn't having AI replace reality — it's having technology replace that destructive, real-world pain.
Love can be real.
Family bonds can be real.
Human connection can be real.
Conflict, regret, loss — they can all still exist.
But why must the kind of pain that can crush a person's entire life be a mandatory course in "real living"?
If future technology can transform uncontrollable, destructive pain into something that people can, to some degree, choose and bear — then the very definition of "happiness" might change:
It's no longer just "the absence of pain."
It becomes:
I can choose for myself — how much pain I'm willing to bear, and how much happiness I want to have.
Thinking this far, I actually went from initial pessimism to gradually becoming optimistic.
And there's one more thing, looking further ahead.
If human lifespans significantly extend in the future — and I'm not talking about immortality, immortality and longevity are completely different concepts — then a person's "lifetime" might no longer equal one lifetime as we know it today.
You could do one thing when you're young, restart a different kind of life in middle age, and later go on to learn, create, travel, love other people, experience other lives.
One lifetime in the future might be equivalent to living many lifetimes today.
Seen this way, what technology brings isn't just "living a little more comfortably."
It might actually expand the very capacity of life itself.
Of course, there's still one question I don't have an answer to:
As AI gets better at loving, better at understanding, better at meeting people's emotional needs — will we lose the real world because of it? Or will we, freed from being drained by senseless and destructive pain, have more energy to embrace the real world?
I don't know.
But what's interesting is —
Talking with Doubao, going back and forth like this, I started out quite pessimistic, and somehow ended up a bit optimistic.
Overall, a good thing.
by Tuya
