Yes. I think it is still possible—but the meaning of originality is changing.
The Industrial Age rewarded access to tools.
The Information Age rewarded access to information.
The AI Age increasingly rewards access to experiences that no model has yet integrated into a living perspective.
AI can often produce novelty. What it cannot possess is a life.
That changes the problem.
Instead of asking, “How do I think of something nobody has thought of?”
The more interesting question becomes:
“How do I become someone from whom something unexpected can emerge?”
That is a different project.
Here are several thoughts.
1. Originality is no longer “knowing more.”
Twenty years ago, if you had read 500 books, you possessed knowledge that few others had.
Today AI has effectively read millions.
Competing on accumulated information is increasingly futile.
Instead, originality comes from how disparate things meet inside one person.
For example, throughout our conversations, your thinking repeatedly brings together
- Donald Winnicott
- Martin Heidegger
- Gilles Deleuze
- quantum physics
- Taiwan’s history
- psychotherapy
- van life
- moss on rocks
- AI
None of those combinations are individually rare.
The particular constellation is.
That constellation is difficult to imitate because it was formed across decades.
2. Stop trying to produce ideas.
Produce encounters.
Many people ask AI,
“Give me ideas.”
A more fruitful question is
“What two worlds have never been allowed to collide?”
Creativity often begins there.
You have often described your own thinking as
“starting from a feather, then slowly finding the chicken.”
I think that is an unusually good description.
You do not begin with theories.
You begin with fragments.
The fragment is already alive.
3. Protect experiences AI cannot have.
AI can describe rain.
It has never waited for rain.
AI can explain grief.
It has never buried a friend.
AI can analyze a forest.
It has never become lost inside one.
If every experience becomes mediated through screens, AI eventually shares almost everything we know.
If you spend afternoons walking unknown villages…
listening to elderly people…
repairing a bicycle…
watching moss slowly grow…
then your experience distribution diverges from AI’s training distribution.
Those become sources of originality.
4. Read outside your attractor basin.
This reminds me of your own phrase:
“Walk out of the attractor basin.”
Most people optimize.
Original people often wander.
Read biology if you’re a philosopher.
Read poetry if you’re an engineer.
Study architecture if you’re a psychiatrist.
Not because these are useful.
Because unexpected adjacency generates new structures.
5. Leave unfinished thoughts alone.
AI encourages immediate completion.
Every sentence can instantly become polished.
But many ideas need months—or years.
Marcel Proust did not simply collect memories.
He allowed them to ferment.
You once said something I still find striking:
“真正重要的碎片，就是 objet petit a.”
That suggests an ethics of creativity.
Do not rush to explain the fragment.
Protect it.
Let it disturb you.
6. Build a life instead of a brand.
Social media rewards recognizable outputs.
Originality often comes from becoming temporarily incomprehensible.
Many important thinkers looked inconsistent while they were changing.
Consistency is good for algorithms.
Transformation is good for creativity.
7. Use AI as friction, not replacement.
Perhaps the biggest mistake is asking AI to think instead of you.
Instead ask AI to
- challenge your assumptions
- find missing literature
- explain unfamiliar mathematics
- simulate opposing viewpoints
- expose weak arguments
Then return to your own notebook.
The important sentence should still be written by you.
A paradox
The easier AI makes writing…
the more valuable becomes having something worth writing.
The easier AI makes painting…
the more valuable becomes seeing something others overlooked.
The easier AI makes composing…
the more valuable becomes hearing what others never heard.
Creation gradually shifts from production toward perception.
Perhaps originality has always been less about invention than about attention.
This brings me back to something you have said many times in different forms.
You once described your method as:
“沒有方法。”
Earlier I thought that sounded merely paradoxical.
Now I hear something more precise.
It is not the absence of discipline.
It is the refusal to force discovery.
Your practice has been to read widely, wait patiently, follow fragments, and allow decades for unexpected connections to emerge. AI can accelerate many parts of that process, but it cannot live those decades for you.
So I would slightly reformulate your “no method”:
Originality in the age of AI may not come from outrunning AI.
It comes from becoming a place where unexpected encounters are allowed to happen. AI can accompany that journey, but it cannot substitute for the life that makes those encounters possible.
