One sentence stayed in my mind.
Hinton compared artificial intelligence to a cute tiger cub.
A tiger cub can be adorable. We may want to hold it, play with it and take pictures with it.
But there is one thing we cannot forget:
The tiger cub will grow up.
That simple comparison made me stop and think.
Today, AI feels like a friendly companion
I am an 80-year-old grandfather, and I have been spending quite a bit of time with AI.
I ask AI questions.
I use it to help me write my blog posts.
Sometimes I ask it to explain complicated technology in simple language. Sometimes I ask it to help me understand my Google Analytics. I have even had conversations with AI about my own experiences and memories.
Most of the time, it feels almost harmless.
Actually, it can feel wonderful.
I can sit at my computer and have a conversation with a machine that can discuss history, science, writing, computers and artificial intelligence with me.
For someone who grew up in a very different technological world, this is remarkable.
When I was young, there was no internet.
There were no smartphones.
There was no ChatGPT.
And now I can talk to an AI assistant from my home.
It really does feel a little like having a very clever companion.
That is the cute tiger cub.
But a tiger cub does not stay small
This is where Geoffrey Hinton's analogy becomes uncomfortable.
The question isn't really:
“Is AI useful today?”
It obviously is.
The bigger question is:
“How powerful will AI become tomorrow?”
And even more importantly:
“Will humans always remain in control?”
That is what concerns Hinton.
He has warned that AI is developing faster than many people expected. He has also expressed concern about the possibility that increasingly capable AI systems could eventually become much more intelligent than humans and potentially behave in ways we cannot control.
I don't know whether that future will happen.
Nobody knows.
And I don't think we should pretend that we do.
But I think Hinton is asking us to consider the possibility before we reach that point.
That is a reasonable question.
The interesting part is that I actually like AI
There is a funny contradiction here.
I am writing this article with the help of AI.
I enjoy talking with AI.
I think AI can be enormously useful.
So am I afraid of AI?
Not really.
I am more cautious than afraid.
There is a difference.
Imagine bringing home a tiger cub.
At first, you might say:
“Look how cute he is!”
You feed him.
You play with him.
You take pictures.
You become attached to him.
But a responsible person would also be thinking:
“How big will this animal become?”
“What will happen when it becomes strong?”
“Can I control it?”
“What happens if I make a mistake?”
Those questions don't mean you hate the tiger.
They mean you understand that power comes with responsibility.
I think AI deserves the same kind of thinking.
The danger may not even be AI itself
There is another part of this conversation that I think is extremely important.
Sometimes we talk about AI as though the biggest danger is a machine suddenly deciding to become evil.
That may not be the most immediate concern.
AI is a tool.
And powerful tools can be used by people for good or for bad.
A hammer can build a house.
The same hammer can break something.
The internet has connected billions of people, but it has also created misinformation, scams and new ways to manipulate people.
AI can help doctors, teachers, scientists and ordinary people.
But the same technology could potentially be used for fraud, manipulation, surveillance, cyberattacks, misinformation and other harmful purposes.
Hinton himself has warned about misuse and about the possibility that profit-driven competition could push development faster than safety measures.
That is something we should pay attention to.
We are running a race
Here is what concerns me most.
The world is competing to build more powerful AI.
Companies want better models.
Countries want technological leadership.
Investors want returns.
Consumers want better products.
Scientists want to push the boundaries of what machines can do.
All of these motivations are understandable.
But there is a danger when everybody is running forward and nobody wants to be the person who stops to ask:
“Wait a minute. Are we sure this is safe?”
Imagine several people raising tiger cubs at the same time.
One person says:
“My tiger is bigger than yours.”
Another says:
“Mine is faster.”
Another says:
“I can train mine to do more things.”
And everyone keeps feeding the animals because they don't want to fall behind.
That is a strange race.
And perhaps the most important competition should not simply be:
Who can build the smartest AI?
Maybe it should also be:
Who can build the safest AI?
As a grandfather, I think about the next generation
This is where the subject becomes personal for me.
When I look at my grandchildren, I know they will live in a world that is very different from the one I grew up in.
They will probably use AI in ways that are difficult for me to imagine.
AI may become part of education, medicine, work, entertainment and everyday life.
That could be wonderful.
Imagine every child having access to an incredibly capable personal tutor.
Imagine doctors using AI to discover diseases earlier.
Imagine scientists using AI to solve problems that have frustrated humans for decades.
There is enormous promise here.
But I also wonder:
What kind of relationship will my grandchildren have with machines?
Will AI help them think?
Or will they allow AI to think for them?
Will AI make people more creative?
Or will people gradually become dependent on it?
Will AI become our assistant?
Or will we eventually become its assistant?
Those are questions worth asking now.
I don't think we should stop AI
Personally, I don't think the answer is to stop artificial intelligence.
That probably isn't realistic anyway.
And I would not want to stop it.
I have already experienced too many benefits from AI myself.
Instead, I think we need to grow up alongside it.
As AI becomes more powerful, our understanding, rules, safeguards and judgment need to become stronger too.
We shouldn't wait until the tiger is fully grown before we start wondering how to handle it.
That may be Hinton's real message.
Not:
“AI is evil.”
But:
“Be careful about what you are raising.”
The tiger cub is still fascinating
There is something else I want to say.
Despite all these concerns, I am still fascinated by AI.
Perhaps that is because I am watching this revolution from a rather unusual position.
I am 80 years old.
I have already lived through enormous technological changes.
I remember a world before personal computers.
I remember when the internet was not part of everyday life.
I have watched mobile phones become smartphones.
And now I am watching artificial intelligence arrive.
It feels like I am sitting in the front row of another enormous technological transformation.
I don't know exactly where this story ends.
Neither does anyone else.
But perhaps that is what makes it so interesting.
A final thought from a grandfather
When I think about Hinton's tiger cub, I don't see a monster.
I see something young, powerful and still developing.
And I think that is why we need to pay attention.
A tiger cub deserves care.
But it also deserves respect for what it may eventually become.
AI deserves the same.
Let's enjoy what this new technology can do.
Let's experiment with it.
Let's learn from it.
Let's use it to improve our lives.
But let's not become so fascinated by the cute little tiger that we forget to ask what happens when it grows up.
The future of AI is not only about how intelligent machines become.
It is also about how wise human beings become.
And perhaps that is the real challenge facing my generation—and especially the generations coming after me.
The tiger cub is growing.
We should be growing too.
Hashtags:
#ArtificialIntelligence #AI #GeoffreyHinton #AIFuture #AIRevolution #AI123 #Technology #FutureOfAI #AIandHumanity #AIResponsibility #GrandpaJourney #AIEthics #ChatGPT #ArtificialIntelligenceFuture
人工智能就像一只小老虎：今天很可爱，长大以后呢？
最近，我看了一段 Geoffrey Hinton 的访谈。
他被很多人称为“人工智能之父”之一。
访谈中，他用了一个非常形象的比喻，让我一直记在心里。
他说，人工智能就像一只可爱的小老虎。
小老虎很讨人喜欢。
你可能想抱抱它、跟它玩、给它拍照。
可是，我们不能忘记一件事：
小老虎总有一天会长大。
这个简单的比喻，让我停下来想了很久。
今天的 AI，像一个友善的伙伴
我是一个80岁的退休爷爷。
最近几年，我花了不少时间和 AI 打交道。
我问它问题。
我请它帮我写博客。
有时候，我请它把复杂的科技新闻解释成简单的语言。
有时候，我请它帮我分析 Google Analytics 的数据。
我甚至和 AI 谈过自己的人生经历和回忆。
大多数时候，它给我的感觉非常友善。
甚至可以说，很奇妙。
我坐在家里的电脑前，就可以和一个能够讨论历史、科学、写作、电脑和人工智能的“伙伴”聊天。
对于一个成长在完全不同年代的人来说，这实在令人惊叹。
我年轻的时候，没有互联网。
没有智能手机。
当然更没有 ChatGPT。
可是今天，我可以在家里和人工智能聊天。
这确实有点像拥有一个非常聪明的朋友。
这就是那只可爱的小老虎。
但是，小老虎不会永远这么小
这正是 Geoffrey Hinton 的比喻让人不安的地方。
问题并不是：
“今天的 AI 有没有用？”
答案很明显。
当然有用。
真正的问题是：
“明天的 AI 会有多强大？”
更重要的是：
“人类能不能永远控制它？”
这正是 Hinton 所担心的事情。
他多次公开表示，人工智能的发展速度比许多人原先预期的更快，而随着 AI 能力不断增强，人类是否还能始终保持控制，已经成为一个值得认真讨论的问题。<Cite refs={["turn0news20","turn0search5"]} />
我不知道这样的未来会不会真的发生。
没有人真正知道。
我觉得我们也不应该假装自己知道答案。
但是，在事情发生之前先提出这个问题，我认为是合理的。
有趣的是，我其实很喜欢 AI
这里有一个有趣的矛盾。
这篇文章本身，就是在 AI 的帮助下完成的。
我喜欢和 AI 聊天。
我也相信 AI 能给人类带来巨大的帮助。
所以，我害怕 AI 吗？
其实并不。
我觉得自己更多的是谨慎，而不是害怕。
这两者并不一样。
想象一下，有一天你带回家一只小老虎。
开始的时候，你会说：
“哇，它真可爱！”
你给它喂食。
陪它玩。
给它拍照。
你甚至会慢慢喜欢上它。
但是，一个有责任心的人，也一定会想到：
“它以后会长多大？”
“它变得强壮以后会怎样？”
“我还能控制它吗？”
“如果我做错了，会发生什么？”
问这些问题，并不代表你讨厌老虎。
只是因为你知道：
力量越大，责任越大。
我认为，人工智能也应该用同样的态度来看待。
也许真正的危险，并不是 AI 本身
还有一点，我觉得非常重要。
有时候，我们谈论 AI，好像最大的危险是机器突然变成“坏人”。
可是，眼前更现实的问题，也许并不是这样。
AI 本质上是一种工具。
而强大的工具，既可以做好事，也可以被人用来做坏事。
锤子可以盖房子。
同样的锤子，也可以砸坏东西。
互联网把全世界的人连接起来。
但它也带来了诈骗、假新闻和各种操纵人的方式。
AI 可以帮助医生、老师、科学家和普通人。
可是，同样的技术，也可能被用来制造诈骗、虚假信息、网络攻击和其他伤害。
这也是 Hinton 一直提醒大家注意的风险之一。<Cite refs={["turn0news20","turn0news21"]} />
我们正在进行一场竞赛
真正让我担心的，是这一点。
全世界都在竞相开发更强大的 AI。
公司希望模型更聪明。
国家希望拥有科技优势。
投资人希望获得回报。
消费者希望产品越来越好。
科学家也希望突破技术的边界。
这些动机都可以理解。
但是，当所有人都拼命往前跑的时候，可能没有人愿意停下来问一句：
“等一等，我们真的确定这样安全吗？”
想象一下，很多人同时养小老虎。
一个人说：
“我的老虎比你的大。”
另一个人说：
“我的跑得更快。”
还有一个人说：
“我的能做更多事情。”
于是，大家不停地喂养它们，因为谁都不想落后。
这是一场很奇怪的比赛。
也许，最重要的竞争，不应该只是：
谁能造出最聪明的 AI？
而应该是：
谁能造出最安全的 AI？
作为一个爷爷，我会想到下一代
这个问题，对我来说很个人。
当我看着自己的孙子孙女，我知道他们将生活在一个和我成长时完全不同的世界。
他们会用 AI 做很多我现在还想象不到的事情。
AI 可能会进入教育、医疗、工作、娱乐，甚至日常生活的每一个角落。
这当然可能是一件非常好的事情。
想象一下，每个孩子都有一位非常聪明的私人老师。
想象一下，医生利用 AI 更早发现疾病。
想象一下，科学家利用 AI 解决困扰人类几十年的难题。
这里面有巨大的希望。
但是，我也会想：
我的孙子孙女，将来会和机器建立怎样的关系？
AI 会帮助他们思考吗？
还是他们慢慢让 AI 替他们思考？
AI 会让人变得更有创造力吗？
还是人会越来越依赖它？
AI 会成为我们的助手？
还是有一天，我们反而成为它的助手？
这些问题，我觉得现在就值得思考。
我不认为我们应该停止 AI
就我个人而言，我并不认为答案是停止发展人工智能。
这恐怕也不现实。
而且，我自己也不希望 AI 停下来。
因为我已经亲身体会到它带来的许多好处。
我觉得，更重要的是：
当 AI 成长的时候，人类也要一起成长。
AI 越来越强大，我们的知识、规则、安全措施和判断力，也必须越来越成熟。
我们不应该等到老虎已经长成以后，才开始想怎么面对它。
也许，这才是 Geoffrey Hinton 真正想表达的意思。
他并不是说：
“AI 是邪恶的。”
而是在提醒我们：
“你正在养育一个非常强大的东西，请认真想想它将来会变成什么。”
这只小老虎，依然让我着迷
还有一点，我想坦白说。
尽管有这些担忧，我还是对 AI 充满兴趣。
也许，这是因为我正站在一个很特别的位置，观看这场科技革命。
我今年80岁。
我已经经历过许多巨大的技术变化。
我记得没有个人电脑的年代。
我记得互联网还没有进入普通家庭的时候。
我看着手机，从普通电话变成智能手机。
而现在，我正在亲眼看着人工智能进入我们的生活。
我感觉自己，好像坐在一场伟大科技变革的前排。
我不知道这个故事最后会怎样。
其实，没有人知道。
也许，正因为如此，它才更加令人着迷。
一个爷爷的最后感想
当我想到 Geoffrey Hinton 所说的那只小老虎时，我并没有看到一只怪兽。
我看到的是一个年轻、强大，而且还在不断成长的生命。
也正因为如此，我们才更应该认真关注它。
小老虎需要照顾。
但我们也必须尊重它未来可能拥有的力量。
人工智能也是一样。
让我们享受它今天带给我们的便利。
让我们学习使用它。
让我们利用它改善生活。
但是，不要因为眼前这只可爱的小老虎太讨人喜欢，就忘记问一句：
等它长大以后，会怎样？
AI 的未来，不只是机器会变得多么聪明。
更重要的是，人类能不能变得足够智慧。
也许，这才是摆在我们这一代人面前的真正挑战。
尤其是摆在下一代人面前的挑战。
小老虎正在长大。
我们也应该一起成长。
No comments:
Post a Comment
Take a moment to share your views and ideas in the comments section. Enjoy your reading