Name: "人工智能教父 "傑弗裡-辛頓：60 分鐘訪談 ("Godfather of AI" Geoffrey Hinton: The 60 Minutes Interview)
Uploaded: 2025-08-03T01:17:08.000Z
Duration: 13 min 12 s

Whether you think artificial intelligence will save the world or end it, you have Geoffrey Hinton to thank.

無論你認為人工智能將拯救世界還是終結世界，你都要感謝傑弗裡-辛頓。

Hinton has been called the godfather of AI, a British computer scientist whose controversial ideas helped make advanced artificial intelligence possible and so change the world.

辛頓被稱為人工智能教父，他是一位英國計算機科學家，其備受爭議的觀點幫助實現了先進的人工智能，從而改變了世界。

Hinton believes that AI will do enormous good, but tonight he has a warning.

辛頓相信人工智能將帶來巨大的好處，但今晚他要發出警告。

He says that AI systems may be more intelligent than we know and there's a chance the machines could take over, which made us ask the question.

他說，人工智能系統可能比我們所知道的更加智能，機器有可能接管一切，這讓我們不禁提出了一個問題。

I think we're moving into a period when, for the first time ever, we may have things more intelligent than us.

我認為，我們正在進入一個時期，有史以來第一次，我們可能擁有比我們更聰明的東西。

You believe these systems have experiences of their own and can make decisions based on those experiences?

你相信這些系統有自己的經驗，並能根據這些經驗做出決定？

I think they probably don't have much self-awareness at present.

我想，他們目前可能還沒有什麼自知之明。

So in that sense, I don't think they're conscious.

是以，從這個意義上說，我不認為他們是有意識的。

And so human beings will be the second most intelligent beings on the planet?

人類將成為地球上第二聰明的生物？

Jeffrey Hinton told us the artificial intelligence he set in motion was an accident born of a  In the 1970s at the University of Edinburgh, he dreamed of simulating a neural network on a computer simply as a tool for what he was really studying, the human brain.

傑弗裡-辛頓（Jeffrey Hinton）告訴我們，他啟動的人工智能是一個意外。 20 世紀 70 年代，在愛丁堡大學，他夢想在計算機上模擬一個神經網絡，作為他真正研究人腦的工具。

But back then, almost no one thought software could mimic the brain.

但在當時，幾乎沒有人認為軟件可以模擬大腦。

His PhD advisor told him to drop it before it ruined his career.

他的博士生導師讓他放棄，免得毀了自己的前途。

Hinton says he failed to figure out the human mind, but the long pursuit led to an artificial version.

辛頓說，他沒能弄清人類的思維，但經過長期的探索，他終於找到了人工智能。

It took much, much longer than I expected.

花的時間比我預想的要長很多很多。

It took like 50 years before it worked well.

大概過了 50 年，才有了良好的效果。

At what point did you realize that you were right about neural networks and most everyone else was wrong?

你是在什麼時候意識到自己對神經網絡的看法是正確的，而其他人都錯了？

In 2019, Hinton and collaborators Jan LeCun on the left and Yoshua Bengio won the Turing Award, the Nobel Prize of Computing.

2019 年，Hinton 和合作者 Jan LeCun（左）以及 Yoshua Bengio 獲得了計算機領域的諾貝爾獎--圖靈獎。

To understand how their work on artificial neural networks helped machines learn to learn, let us take you to a game.

為了瞭解他們在人工神經網絡方面的研究如何幫助機器學會學習，讓我們帶您去看一場遊戲。

This is Google's AI lab in London, which we first showed you this past April.

這是谷歌位於倫敦的人工智能實驗室，我們在今年四月首次向您展示了該實驗室。

Geoffrey Hinton wasn't involved in this soccer project, but these robots are a great example of machine learning.

傑弗裡-辛頓沒有參與這個足球項目，但這些機器人是機器學習的典範。

The thing to understand is that the robots were not programmed to play soccer.

需要了解的是，機器人的程序並不是用來踢足球的。

一般來說，人工智能是這樣做的。

Hinton and his collaborators created software in layers with each layer handling part of the problem.

辛頓和他的合作者創建了分層軟件，每層處理一部分問題。

When, for example, the robot scores, a message is sent back down through all of the layers that says that pathway was right.

例如，當機器人得分時，所有層都會向下發送一條資訊，說明這條路徑是正確的。

Likewise, when an answer is wrong, that message goes down through the network.

同樣，當答案錯誤時，該資訊也會通過網絡傳遞下去。

So correct connections get stronger, wrong connections get weaker, and by trial and error, the machine teaches itself.

是以，正確的連接會變得更強，錯誤的連接會變得更弱，通過不斷嘗試和錯誤，機器會自我學習。

You think these AI systems are better at learning than the human mind?

你認為這些人工智能系統的學習能力比人類更強？

And at present, they're quite a lot smaller.

而目前，它們的規模要小得多。

So even the biggest chatbots only have about a trillion connections in them.

是以，即使是最大的哈拉機器人，也只有大約一萬億個連接。

The human brain has about a hundred trillion.

And yet, in the trillion connections in a chatbot, it knows far more than you do in your hundred trillion connections, which suggests it's got a much better way of getting knowledge into those connections.

然而，在一個擁有千兆級連結的聊天機器人裡，它所知道的遠比你那擁有百兆級連結的大腦還多，這暗示它有一種更有效率的方式，將知識灌輸到那些連結中。

A much better way of getting knowledge that isn't fully understood.

這是獲取尚未完全理解的知識的更好方法。

We have a very good idea of sort of roughly what it's doing.

我們對它的大致作用有一個很好的瞭解。

But as soon as it gets really complicated, we don't actually know what's going on any more than we know what's going on in your brain.

但一旦變得非常複雜，我們實際上就不知道發生了什麼，就像我們不知道你的大腦裡發生了什麼一樣。

What do you mean we don't know exactly how it works?

你說我們不知道它的具體工作原理是什麼意思？

What we did was we designed the learning algorithm.

我們所做的就是設計學習算法。

That's a bit like designing the principle of evolution.

But when this learning algorithm then interacts with data, it produces complicated neural networks that are good at doing things, but we don't really understand exactly how they do those things.

但是，當這種學習算法與數據交互時，就會產生複雜的神經網絡，這些網絡擅長做一些事情，但我們並不真正瞭解它們究竟是如何做這些事情的。

What are the implications of these systems autonomously writing their own computer code and executing their own computer code?

這些系統自主編寫自己的計算機代碼並執行自己的計算機代碼會產生什麼影響？

So one of the ways in which these systems might escape control is by writing their own computer code to modify themselves.

是以，這些系統可能擺脫控制的方法之一就是編寫自己的計算機代碼來進行自我修改。

And that's something we need to seriously worry about.

這是我們需要認真擔心的問題。

What do you say to someone who might argue, if the systems become malevolent, just turn them off?

有人可能會說，如果系統變得邪惡，把它們關掉不就行了嗎？

They will be able to manipulate people, right?

And these will be very good at convincing people, because they'll have learned from all the novels that were ever written, all the books by Machiavelli, all the political connivances.

這些人非常善於說服別人，因為他們已經從所有的小說、馬基雅維利的所有著作以及所有的政治縱容中吸取了教訓。

They'll know all that stuff, they'll know how to do it.

他們知道所有這些東西，知道如何去做。

Know-how of the human kind runs in Geoffrey Hinton's family.

傑弗裡-辛頓的家族中流傳著人類的訣竅。

His ancestors include mathematician George Boole, who invented the basis of computing, and George Everest, who surveyed India and got that mountain named after him.

他的祖先包括數學家喬治-布爾（George Boole）和喬治-埃弗勒斯（George Everest），前者發明了計算機的基礎，後者勘測了印度並以自己的名字命名了那座山。

But as a boy, Hinton himself could never climb the peak of expectations raised by a domineering father.

但是，作為一個男孩，辛頓自己卻永遠無法攀上專橫父親所寄予的期望的頂峰。

Every morning when I went to school, he'd actually say to me as I walked down the driveway, get in there pitching, and maybe when you're twice as old as me, you'll be half as good.

每天早上我去上學時，他總會對我說：‘上場投球去吧，也許當你年紀是我兩倍時，你能有我一半厲害。

He knew a lot more about beetles than he knew about people.

他對甲蟲的瞭解遠遠多於對人的瞭解。

When he died, we went to his study at the university, and the walls were lined with boxes of papers on different kinds of beetle.

他去世後，我們去了他在大學的書房，牆上掛滿了一箱箱關於各種甲蟲的論文。

And just near the door, there was a slightly smaller box that simply said, not insects.

就在門附近，有一個稍小的盒子，上面只寫著 "非昆蟲"。

And that's where he had all the things about the family.

在那裡，他有關於這個家庭的所有資料。

Today, at 75, Hinton recently retired after what he calls 10 happy years at Google.

如今，75 歲的辛頓在谷歌度過了 10 年快樂的時光，最近剛剛退休。

Now, he's professor emeritus at the University of Toronto, and he happened to mention he has more academic citations than his father.

現在，他是多倫多大學的名譽教授，他偶然提到自己的學術論文引用次數比他父親還多。

Some of his research led to chatbots like Google's Bard, which we met last spring.

他的一些研究成果促成了哈拉機器人的誕生，比如谷歌的 "巴德"（Bard），我們去年春天就見過它。

We asked Bard to write a story from six words.

我們要求巴德用六個單詞寫一個故事。

The shoes were a gift from my wife, but we never had a baby.

這雙鞋是我妻子送給我的禮物，但我們沒有孩子。

Bard created a deeply human tale of a man whose wife could not conceive, and a stranger who accepted the shoes to heal the pain after her miscarriage.

巴德創作了一個深刻的人性故事：一個男人的妻子無法懷孕，而一個陌生人在她流產後接受了她的鞋子，以治癒她的痛苦。

Chatbots are said to be language models that just predict the next most likely word based on probability.

據說，哈拉機器人是一種語言模型，它只是根據概率預測下一個最可能出現的單詞。

You'll hear people saying things like, they're just doing autocomplete.

你會聽到有人說，他們只是在做自動完成。

They're just trying to predict the next word.

Well, it's true they're just trying to predict the next word.

沒錯，他們只是想預測下一個詞。