Name: ChatGPT CEO 山姆阿特曼表示 AI「無法阻擋」 (Sam Altman: there’s no “magic red button” to stop AI)
Uploaded: 2024-01-30T21:01:20.000Z
Duration: 3 min 55 s

These are two of the most important people shaping the future of artificial intelligence.

這兩位是塑造人工智慧未來的兩位最重要的人物。

形容詞最高級

Sam Altman, the CEO of Open AI, the startup behind ChatGPT and Satya Nadella, the CEO of Microsoft, Open AI's biggest investor.

Sam Altman是OpenAI的CEO，該新創公司是ChatGPT背後的推動者，而Satya Nadella是Microsoft的CEO，也是OpenAI最大的投資者。

過去簡單式

They spoke to the Economists editor in chief about what the future of AI really looks like.

他們向《經濟學人》的總編輯談論了人工智慧的真實未來形勢。

Sam, let's start with you. What are the most important capabilities that Chat GPT will develop in the next year?

山姆，讓我們從你開始。ChatGPT在接下來的一年中將發展的最重要的能力是什麼？

The world had like a two week freak out with GPT-4, right?

世界對GPT-4有種像是持續兩週的狂熱反應，對吧？

This changes everything. AGI is coming tomorrow.

這改變一切。通用人工智慧（AGI）就要來了。

There are no jobs by the end of the year, and now people are like, "Why is it so slow?"

年底前沒有新的進展，現在人們都在問：「為什麼這麼慢？」

And I love that, I think that's a great thing about the human spirit that we always want more and better.

我喜歡這一點，我認為這是人類精神的一個很棒的特點，我們總是追求更多和更好。

We've not had this where there has been some general purpose technology whose diffusion happened instantaneously everywhere.

我們還沒有經歷過一種普遍技術，其傳播在所有地方瞬間發生。

In any place, health care and education is most of the government spent.

在任何地方，政府開支的大部分都用於醫療保健和教育。

You now have, right, the ability to give every student and every citizen of the of the globe a better health advice and a better personalized tutor.

現在，有能力為全球的每位學生和每位公民提供更好的健康建議和更好的個性化教學。

You know, I often think one interesting way to measure this is what percent of tasks can GPT-4 do, let's say it's 10%, can GPT-5 do 12% of human tasks or 15% or 20%?

你知道，我經常思考一種有趣的方式來衡量這一點，就是GPT-4可以完成多少任務，比方說它可以完成10%的任務，那麼GPT-5能完成12%、15%還是20%的人類任務呢？

But the fact that so many people are able to use it for productivity in their workflow, that's the power.

但事實上，這麼多人能夠在工作流程中使用它提高生產力，這就是它的威力。

And I know it's not as satisfying as saying, here's each thing it will do, but it's that it becomes a companion for knowledge work.

我知道這並不像說明每件事情它會做那麼令人滿意，但它成為了知識工作的一個伴侶。

And that's a absolutely appropriate answer to my question.

這絕對是對我的問題的一個合適的回答。

But I guess I'm trying to get at the sense of whether this is incremental or whether it's radical, even in the next year.

但我想我嘗試著理解這是一種漸進的變化還是一種激進的變革，即使在接下來的一年中。

I believe that someday we will make something that qualifies as an AGI by whatever fuzzy definition you want. The world will have a two week freak out and then people will go on with their lives.

我相信總有一天我們會創造出符合任何你想要的模糊定義的通用人工智慧（AGI）。世界將會在兩週內發生狂熱，然後人們將繼續他們的生活。

Sam Altman just said the world would only have a two-week freak out when we get to AGI.

Sam Altman剛剛說，當我們達到通用人工智慧（AGI）時，世界只會發生兩週的瘋狂。

One thing I say a lot is no one knows what happens next.

我經常說的一點是沒有人知道接下來會發生什麼。

And I can't see to the other side of that event horizon with any detail, but it does seem like the deep human motivations will not go anywhere.

我無法以任何細節看到那個事件地平線的另一邊，但似乎深層次的人類動機不會消失。

This is when people start getting alarmed that we have no idea. - Why?

這是當人們開始感到驚慌，因為我們一無所知的時候。- 為什麼呢？

Well, I think that's just... - We're gonna habe some intelligence that is more intelligence than all of us and we have no idea what happened.

嗯，我想這只是... - 我們將擁有一種比我們所有人都更聰明的智能，而我們對發生了什麼毫不知情。

No, no, I like one thing I love to do is go back and read about the contemporaneous accounts of technological revolutions at the time and the expert predictions are just always totally wrong.

不，不，我喜歡做的一件事是回顧當時有關技術革命的同時代記錄，專家的預測總是完全錯誤的。

And you need to have some flexibility in your opinions and look, have a tight feedback loop of how it's going with the world.

而且你需要對你的觀點有一些靈活性，看看與世界的相應反饋是如何的。

The amount of focus on safety and regulation is sort of very, very high.

對安全和監管的關注程度非常非常高。

Tell me whether you think regulators have got it right or whether we're not doing enough.

告訴我你是否認為監管機構做對了，還是我們還沒有做到足夠。

At this point, you know, if I look at what the White House EO is or what the UK Safety Summit is, what's happening in Europe, what's happening in Japan,

在這一點上，你知道，如果我看看白宮的執行令，或者英國的安全峰會是什麼，歐洲正在發生什麼，日本正在發生什麼，

they are going to have a say, nation states are absolutely going to have a say on what is the regulation that controls any technology development.

他們將發表意見，國家絕對將對控制任何技術發展的規定發表意見。

Most importantly, what is ready for deployment or not.

最重要的是，什麼是準備好投入使用的，什麼還不是。

And so I feel like we will all be subject to those rags.

因此，我覺得我們都將受到這些法規的約束。

If you were ever in that room and you thought to yourself, "This is getting dangerous and this could actually have consequences that I would not want upon the world."

如果你曾經在那個房間裡，並且你心裡想，「這變得很危險，這實際上可能會產生我不希望看到的後果。」

Would you then shout stop and would you stop?

那麼你會大聲呼喊停下來嗎？你會停下來嗎？

There's no, like, one big magic red button we have that blows up the data center which I think some people sort of assume exists.

這件事必沒有一個大魔法紅色按鈕，我們可以按下它，炸毀數據中心，這是一些人可能會假設存在的。

這不是一個二元的啟動-停止的決定。

It is the many little decisions along the way about allow this, don't allow this.

這是沿途的許多小決策，例如允許這個，不允許那個。

How to predict what the the risks of the future are going to be, how to mitigate those set this new value here.

如何預測未來的風險，如何減輕這些風險，如何設定這個新的價值。

Thanks for watching. To watch the full interview, click the link and subscribe to the Economist.

感謝觀看。要觀看完整的訪談，請點擊連結並訂閱《經濟學人》。