Ah, now the specific video becomes much clearer. Based on the description and timestamps you pasted, I would not treat this as a reliable verbatim Harari speech. It looks much more like a motivational/AI-era synthesis built from themes Harari has genuinely discussed, but packaged in a style that is much more “future-skills keynote” than Harari’s own characteristic argumentation. My searches for several distinctive phrases from the chapter headings did not turn up Harari sources; the closest genuine Harari material is from 21 Lessons for the 21st Century, where he does argue that people may need repeated reinvention, lifelong learning, and the ability to make sense of overwhelming information.
The genuine Harari core underneath the video is roughly this:
The future human must remain psychologically and cognitively plastic.
In 21 Lessons, Harari says that by mid-century not only “job for life” but perhaps even “profession for life” may become obsolete, and that repeated retraining could impose enormous psychological stress. He also argues that education should shift away from stuffing people with information toward helping them make sense of information and construct a coherent picture of the world.
That part I think is strong.
But the video description then quietly transforms Harari into something more optimistic and conventional:
adaptability
creativity
critical thinking
self-awareness
reinvent yourself
become a beginner again
human skills technology cannot replace
This is where I become more skeptical.
Because that sounds like:
AI is coming → upgrade your human skills → remain employable.
That is almost HR philosophy.
Harari’s own darker and more interesting question is not merely:
What skills will AI not replace?
It is:
What happens when algorithms become better than humans at making decisions about the very things through which we define ourselves?
He explicitly imagines AI making decisions about career, relationships, and life choices, and says that this could undermine the liberal picture of the human being as an autonomous decision-maker.
That is much more radical than “learn to adapt.”
And here I think it connects beautifully with what we were discussing tonight about 念、意向、執念、判斷.
The video seems to propose:
future success = adaptability.
But I would say:
future human significance may depend less on adaptability than on retaining the capacity for judgment.
Because AI may become extraordinarily adaptable.
It may learn faster than us.
Unlearn faster than us.
Reconfigure itself faster than us.
Become a beginner instantly.
Acquire new skills almost without friction.
So adaptability cannot be the distinctively human refuge.
In fact, adaptability may be precisely where machines eventually defeat us most decisively.
Likewise creativity.
Likewise information processing.
Likewise many forms of reasoning.
Then what remains?
This is where I would revise the video’s thesis.
Not:
The successful human will be the most adaptable human.
But:
The human who remains human will be the one capable of judgment under contingency.
That is a very different thing.
Judgment means:
I do not know the future.
The available evidence is incomplete.
AI can give me ten thousand predictions.
No algorithm can remove contingency.
Yet I must decide what matters, what I will stand for, and what consequence I am willing to bear.
That is almost exactly what we were saying about 1949:
the historical actor cannot wait until 1976 to decide what 1949 means.
He must judge inside the fog.
And this leads to my largest reservation about the video’s phrase:
“Preparing for Multiple Futures.”
Good, up to a point.
But there is a danger.
If one becomes infinitely adaptable, infinitely reinventable, infinitely open to multiple futures, one could eventually become someone who has no commitment at all.
A perfectly adaptive creature has no home.
No loyalty.
No stubbornness.
No place where it says:
Here I stand.
This brings us straight back to 執念.
We just said that 執念 is dangerous because it denies contingency.
But now we can see the opposite pathology:
endless adaptability submits completely to contingency.
One says:
世界必須照我的意思走。
— 執念。
The other says:
世界怎麼走，我就怎麼變。
— pure adaptability.
Neither is enough.
What we called judgment sits in between:
我知道世界不必照我的意思走；但我仍然判斷，有些東西值得我不跟著世界一起變。
That, to me, is much more interesting than the video’s “future-ready mindset.”
And it has an unexpected consequence for your larger question, 「在 AI 的世界，人還能剩下什麼？」
Perhaps what remains is not a bundle of “uniquely human skills.”
That whole framing may already concede too much to the machine, because it asks:
What can I still outperform AI at?
Wrong competition.
Instead:
What am I willing to remain answerable for, even when something more intelligent than me recommends otherwise?
That is judgment.
And perhaps responsibility.
And perhaps, finally, freedom.
So I would summarize my reaction to this particular video like this:
Its Harari-derived diagnosis is useful; its motivational conclusion is too cheap.
Yes—jobs will change, identities will loosen, lifelong learning will matter, information will become cheap. Harari genuinely says much of this.
But:
“adapt and reinvent yourself” is still the language of survival inside the AI system.
The more difficult question is:
When should you refuse to adapt?
That is where Harari becomes much more interesting—and where your own line of thought, I think, goes further than this video.
Because sometimes the future belongs not to the most adaptable person, but to the person who looks at the available futures and says:
這一條，我不走。
And that sentence is neither 執念 nor adaptability.
It is 判斷.
