個(gè)人的品味有時(shí)候是可以預(yù)測(cè)的。青少年為了炫耀自己的閱讀習(xí)慣,通常會(huì)選擇《了不起的蓋茨比》(The Great Gatsby)、《麥田里的守望者》(Catcher in The Rye)和《1984》等經(jīng)典文學(xué)作品。但埃隆·馬斯克早在成為一位與眾不同的首席執(zhí)行官之前,就在努力打破常規(guī),他不喜歡常見的經(jīng)典文學(xué)作品,而是更喜歡著名的科幻小說。這讓他在同學(xué)當(dāng)中顯得格格不入,但對(duì)于創(chuàng)建了SpaceX和OpenAI的馬斯克而言,他的這種行為在預(yù)料之中。
馬斯克發(fā)現(xiàn),科幻作品以宗教和單純的科學(xué)做不到的方式,在他的青少年時(shí)期培養(yǎng)了他的好奇心。沃爾特·艾薩克森在他的新傳記《埃隆·馬斯克》(Elon Musk)里寫道,對(duì)于他思考的宇宙如何起源和宇宙為什么存在這些問題,宗教和科學(xué)無法提供令他滿意的答案。
艾薩克森寫道:“到了青少年時(shí)期,馬斯克覺得若有所失,這讓他感到困擾。”這種感覺讓他產(chǎn)生了“青春期存在危機(jī)”,馬斯克試圖通過看書來解決危機(jī)。
馬斯克說:“我開始努力探尋生命和宇宙的意義。但我變得非常沮喪,因?yàn)樯赡軟]有任何意義。”他讀了尼采等存在主義哲學(xué)家的書,但他們卻讓他感覺更加迷茫(他不推薦將這類書籍作為青少年閱讀材料)。相反,他在超自然的世界里找到了安慰。
艾薩克森寫道,有三位作者和他們的作品指引他度過了那個(gè)階段,并啟發(fā)他產(chǎn)生了移民火星和創(chuàng)造有益于而非有害于人類的機(jī)器人這種想法。
《嚴(yán)厲的月亮》(The Moon Is a Harsh Mistress),作者:羅伯特·海因萊因
這是馬斯克最喜歡的一本小說。故事發(fā)生在由一臺(tái)名為邁克的超級(jí)計(jì)算機(jī)管理的月球流放地。艾薩克森寫道,這臺(tái)人工智能超越了機(jī)器人的狀態(tài),因?yàn)椤八凶灾饕庾R(shí)和幽默感”,這讓它在一場(chǎng)叛亂中選擇了自我犧牲。
艾薩克森解釋稱:“這本書所探討的問題后來成為馬斯克人生的核心:人工智能的發(fā)展會(huì)有益于人類并且能夠保護(hù)人類,還是機(jī)器會(huì)產(chǎn)生自主動(dòng)機(jī),變成人類的威脅。”
當(dāng)馬斯克在2015年與薩姆·奧爾特曼合作創(chuàng)建OpenAI時(shí),他很快就遭遇了這個(gè)問題。他經(jīng)常會(huì)談?wù)撊斯ぶ悄艿奈kU(xiǎn),他認(rèn)為人工智能系統(tǒng)需要安全保障措施,以避免它們?nèi)〈祟怺他經(jīng)常與谷歌(Google)的聯(lián)合創(chuàng)始人拉里·佩奇辯論這個(gè)問題,這最終導(dǎo)致兩人關(guān)系破裂]。馬斯克對(duì)人工智能安全性的立場(chǎng)奠定了OpenAI經(jīng)營(yíng)目標(biāo)的基礎(chǔ)。OpenAI網(wǎng)站上的一份聲明顯示,其目標(biāo)是“開發(fā)最有可能造福全體人類的數(shù)字智能”。
艾薩克·阿西莫夫的《機(jī)器人》(Robot)系列
馬斯克從艾薩克·阿西莫夫圍繞機(jī)器人創(chuàng)作的37篇短篇小說和六本長(zhǎng)篇小說里,找到了發(fā)展未來人工智能的更多靈感。與海因萊因的作品一樣,阿西莫夫的作品同樣探討了人工智能道德準(zhǔn)則這個(gè)話題。艾薩克森寫道:“這些小說制定的機(jī)器人定律,旨在保證機(jī)器人不會(huì)失控。”
為了不讓讀者感到枯燥無聊,阿西莫夫在一本書中把一條規(guī)則命名為“零號(hào)定律”(Zeroth Law),內(nèi)容是“機(jī)器人不得傷害人類,或者通過不作為使人類受到傷害”。阿西莫夫的這些作品產(chǎn)生了深遠(yuǎn)影響:正如艾薩克森所說,馬斯克在幾十年后發(fā)推文稱“《基地》系列(Foundation Series)和零號(hào)定律對(duì)于SpaceX的創(chuàng)立至關(guān)重要。”
馬斯克與奧爾特曼創(chuàng)立OpenAI,希望以有益于人類的方式開發(fā)人工智能,在這個(gè)過程里,阿西莫夫的作品也產(chǎn)生了影響。“人工智能對(duì)齊”問題旨在使通用人工智能(AGI)與人類的價(jià)值觀和意圖保持一致,這與阿西莫夫在小說中提出的旨在阻止機(jī)器人控制人類的法則,有異曲同工之妙。
《銀河系漫游指南》(The Hitchhiker’s Guide to the Galaxy),作者:道格拉斯·亞當(dāng)斯
艾薩克森將這本經(jīng)典科幻作品列為對(duì)馬斯克的“純真年代”影響最大的一本小說。他寫道,這本小說“幫助塑造了馬斯克的哲學(xué)觀,并讓他在嚴(yán)肅之外多了一些滑稽的幽默感”。亞當(dāng)斯沒有像馬斯克一樣開一些露骨的笑話,而是采用了諷刺的描寫方法,馬斯克稱這幫助他擺脫了抑郁的狀態(tài)。
這本書的核心主題同樣是理解人生的意義。在書里,人們要求超級(jí)計(jì)算機(jī)回答宇宙的目的是什么,700萬年后,計(jì)算機(jī)回答了“42”。計(jì)算機(jī)對(duì)震驚的人們說道:“坦率地說,問題在于你們從來都沒有真正知道問題是什么。”
馬斯克似乎將他從這個(gè)宇宙中得到的啟示,以及后來對(duì)視頻游戲的興趣,融入到了他目前的每一次創(chuàng)業(yè)當(dāng)中。馬斯克說:“這本書讓我認(rèn)識(shí)到,我們需要擴(kuò)大科學(xué)的范圍,以便于更好地提出關(guān)于宇宙的問題。”他在創(chuàng)建自己的人工智能公司xAI時(shí)也貫徹了這種思想。2018年,由于奧爾特曼拒絕了他的公司經(jīng)營(yíng)方案,馬斯克離開OpenAI,創(chuàng)建了xAI。
他在7月的一次Twitter Spaces Talk中表示,xAI旨在“打造一款優(yōu)秀的通用人工智能,其根本目的是嘗試了解宇宙。開發(fā)人工智能最安全的方法實(shí)際上是創(chuàng)造擁有最大限度好奇心和追求真相精神的人工智能。”
今年早些時(shí)候,投資者為該領(lǐng)域的技術(shù)進(jìn)步投入了數(shù)十億美元,人工智能創(chuàng)新可能開始讓人感覺不可思議,并且讓人們不安地回想起一些著名的科幻小說。上周,在有關(guān)這個(gè)話題的一次峰會(huì)上,馬斯克引用他最喜歡的書中的內(nèi)容,對(duì)美國(guó)參議員稱人工智能是一把雙刃劍,并請(qǐng)求監(jiān)管深度人工智能。
作為人工智能領(lǐng)域的早期投資者之一,馬斯克希望避免出現(xiàn)他研究過的那些科幻設(shè)定,即一位以自我為中心的科學(xué)家某一天對(duì)自己的發(fā)明失去控制。但他已經(jīng)后悔離開OpenAI,他說自己那樣做是“極其愚蠢的”。或許窩在沙發(fā)里讀一本書,可以緩解他的這種情緒。(財(cái)富中文網(wǎng))
譯者:劉進(jìn)龍
審校:汪皓
個(gè)人的品味有時(shí)候是可以預(yù)測(cè)的。青少年為了炫耀自己的閱讀習(xí)慣,通常會(huì)選擇《了不起的蓋茨比》(The Great Gatsby)、《麥田里的守望者》(Catcher in The Rye)和《1984》等經(jīng)典文學(xué)作品。但埃隆·馬斯克早在成為一位與眾不同的首席執(zhí)行官之前,就在努力打破常規(guī),他不喜歡常見的經(jīng)典文學(xué)作品,而是更喜歡著名的科幻小說。這讓他在同學(xué)當(dāng)中顯得格格不入,但對(duì)于創(chuàng)建了SpaceX和OpenAI的馬斯克而言,他的這種行為在預(yù)料之中。
馬斯克發(fā)現(xiàn),科幻作品以宗教和單純的科學(xué)做不到的方式,在他的青少年時(shí)期培養(yǎng)了他的好奇心。沃爾特·艾薩克森在他的新傳記《埃隆·馬斯克》(Elon Musk)里寫道,對(duì)于他思考的宇宙如何起源和宇宙為什么存在這些問題,宗教和科學(xué)無法提供令他滿意的答案。
艾薩克森寫道:“到了青少年時(shí)期,馬斯克覺得若有所失,這讓他感到困擾。”這種感覺讓他產(chǎn)生了“青春期存在危機(jī)”,馬斯克試圖通過看書來解決危機(jī)。
馬斯克說:“我開始努力探尋生命和宇宙的意義。但我變得非常沮喪,因?yàn)樯赡軟]有任何意義。”他讀了尼采等存在主義哲學(xué)家的書,但他們卻讓他感覺更加迷茫(他不推薦將這類書籍作為青少年閱讀材料)。相反,他在超自然的世界里找到了安慰。
艾薩克森寫道,有三位作者和他們的作品指引他度過了那個(gè)階段,并啟發(fā)他產(chǎn)生了移民火星和創(chuàng)造有益于而非有害于人類的機(jī)器人這種想法。
《嚴(yán)厲的月亮》(The Moon Is a Harsh Mistress),作者:羅伯特·海因萊因
這是馬斯克最喜歡的一本小說。故事發(fā)生在由一臺(tái)名為邁克的超級(jí)計(jì)算機(jī)管理的月球流放地。艾薩克森寫道,這臺(tái)人工智能超越了機(jī)器人的狀態(tài),因?yàn)椤八凶灾饕庾R(shí)和幽默感”,這讓它在一場(chǎng)叛亂中選擇了自我犧牲。
艾薩克森解釋稱:“這本書所探討的問題后來成為馬斯克人生的核心:人工智能的發(fā)展會(huì)有益于人類并且能夠保護(hù)人類,還是機(jī)器會(huì)產(chǎn)生自主動(dòng)機(jī),變成人類的威脅。”
當(dāng)馬斯克在2015年與薩姆·奧爾特曼合作創(chuàng)建OpenAI時(shí),他很快就遭遇了這個(gè)問題。他經(jīng)常會(huì)談?wù)撊斯ぶ悄艿奈kU(xiǎn),他認(rèn)為人工智能系統(tǒng)需要安全保障措施,以避免它們?nèi)〈祟怺他經(jīng)常與谷歌(Google)的聯(lián)合創(chuàng)始人拉里·佩奇辯論這個(gè)問題,這最終導(dǎo)致兩人關(guān)系破裂]。馬斯克對(duì)人工智能安全性的立場(chǎng)奠定了OpenAI經(jīng)營(yíng)目標(biāo)的基礎(chǔ)。OpenAI網(wǎng)站上的一份聲明顯示,其目標(biāo)是“開發(fā)最有可能造福全體人類的數(shù)字智能”。
艾薩克·阿西莫夫的《機(jī)器人》(Robot)系列
馬斯克從艾薩克·阿西莫夫圍繞機(jī)器人創(chuàng)作的37篇短篇小說和六本長(zhǎng)篇小說里,找到了發(fā)展未來人工智能的更多靈感。與海因萊因的作品一樣,阿西莫夫的作品同樣探討了人工智能道德準(zhǔn)則這個(gè)話題。艾薩克森寫道:“這些小說制定的機(jī)器人定律,旨在保證機(jī)器人不會(huì)失控。”
為了不讓讀者感到枯燥無聊,阿西莫夫在一本書中把一條規(guī)則命名為“零號(hào)定律”(Zeroth Law),內(nèi)容是“機(jī)器人不得傷害人類,或者通過不作為使人類受到傷害”。阿西莫夫的這些作品產(chǎn)生了深遠(yuǎn)影響:正如艾薩克森所說,馬斯克在幾十年后發(fā)推文稱“《基地》系列(Foundation Series)和零號(hào)定律對(duì)于SpaceX的創(chuàng)立至關(guān)重要。”
馬斯克與奧爾特曼創(chuàng)立OpenAI,希望以有益于人類的方式開發(fā)人工智能,在這個(gè)過程里,阿西莫夫的作品也產(chǎn)生了影響。“人工智能對(duì)齊”問題旨在使通用人工智能(AGI)與人類的價(jià)值觀和意圖保持一致,這與阿西莫夫在小說中提出的旨在阻止機(jī)器人控制人類的法則,有異曲同工之妙。
《銀河系漫游指南》(The Hitchhiker’s Guide to the Galaxy),作者:道格拉斯·亞當(dāng)斯
艾薩克森將這本經(jīng)典科幻作品列為對(duì)馬斯克的“純真年代”影響最大的一本小說。他寫道,這本小說“幫助塑造了馬斯克的哲學(xué)觀,并讓他在嚴(yán)肅之外多了一些滑稽的幽默感”。亞當(dāng)斯沒有像馬斯克一樣開一些露骨的笑話,而是采用了諷刺的描寫方法,馬斯克稱這幫助他擺脫了抑郁的狀態(tài)。
這本書的核心主題同樣是理解人生的意義。在書里,人們要求超級(jí)計(jì)算機(jī)回答宇宙的目的是什么,700萬年后,計(jì)算機(jī)回答了“42”。計(jì)算機(jī)對(duì)震驚的人們說道:“坦率地說,問題在于你們從來都沒有真正知道問題是什么。”
馬斯克似乎將他從這個(gè)宇宙中得到的啟示,以及后來對(duì)視頻游戲的興趣,融入到了他目前的每一次創(chuàng)業(yè)當(dāng)中。馬斯克說:“這本書讓我認(rèn)識(shí)到,我們需要擴(kuò)大科學(xué)的范圍,以便于更好地提出關(guān)于宇宙的問題。”他在創(chuàng)建自己的人工智能公司xAI時(shí)也貫徹了這種思想。2018年,由于奧爾特曼拒絕了他的公司經(jīng)營(yíng)方案,馬斯克離開OpenAI,創(chuàng)建了xAI。
他在7月的一次Twitter Spaces Talk中表示,xAI旨在“打造一款優(yōu)秀的通用人工智能,其根本目的是嘗試了解宇宙。開發(fā)人工智能最安全的方法實(shí)際上是創(chuàng)造擁有最大限度好奇心和追求真相精神的人工智能。”
今年早些時(shí)候,投資者為該領(lǐng)域的技術(shù)進(jìn)步投入了數(shù)十億美元,人工智能創(chuàng)新可能開始讓人感覺不可思議,并且讓人們不安地回想起一些著名的科幻小說。上周,在有關(guān)這個(gè)話題的一次峰會(huì)上,馬斯克引用他最喜歡的書中的內(nèi)容,對(duì)美國(guó)參議員稱人工智能是一把雙刃劍,并請(qǐng)求監(jiān)管深度人工智能。
作為人工智能領(lǐng)域的早期投資者之一,馬斯克希望避免出現(xiàn)他研究過的那些科幻設(shè)定,即一位以自我為中心的科學(xué)家某一天對(duì)自己的發(fā)明失去控制。但他已經(jīng)后悔離開OpenAI,他說自己那樣做是“極其愚蠢的”。或許窩在沙發(fā)里讀一本書,可以緩解他的這種情緒。(財(cái)富中文網(wǎng))
譯者:劉進(jìn)龍
審校:汪皓
Sometimes, taste can be predictable. The pantheon for teenagers trying to flex their reading habits typically includes The Great Gatsby, Catcher in The Rye, or 1984. But even before becoming a CEO who tries to position himself apart from other CEOs, Elon Musk was trying to break from the mold, shirking the usual classics in favor of famous sci-fi lit. It differentiated him from his peers, but was nonetheless a predictable move for the man behind SpaceX and OpenAI.
Musk found that sci-fi addressed a curiosity brewing during his teen years in a way that religion and science alone couldn’t. Neither gave satisfactory answers to his questions about where the universe came from and why it exists, Walter Isaacson wrote in his new biography of the CEO, titled Elon Musk.
“When he reached his teens, it began to gnaw at him that something was missing,” Isaacson wrote. That feeling led to an “adolescent existential crisis” that Musk tried to solve with books.
“I began trying to figure out what the meaning of life and the universe was,” Musk said. “And I got real depressed about it, like maybe life may have no meaning.” He turned to existential philosophers like Nietzsche, but they left him feeling more adrift (he doesn’t recommend this as teenage reading material). He found solace in the supernatural world instead.
Three authors and books, Isaacson wrote, guided Musk through this phase and to the other side of trying to colonize Mars and bringing robots to life in a way that benefits, rather than harms, humanity.
The Moon Is a Harsh Mistress by Robert Heinlein
A favorite of Musk’s, this novel takes place on a lunar penal colony ruled by a supercomputer called Mike. The AI surpasses its robot state “with self-awareness and a sense of humor,” Isaacson writes, which leads to its self-sacrifice during a rebellion.
“The book explores an issue that would become central to Musk’s life: Will artificial intelligence develop in ways that benefit and protect humanity, or will machines develop intentions of their own and become a threat to humans,” Isaacson explains.
Musk would soon grapple with this question when he helped found OpenAI in 2015 alongside Sam Altman. He has often spoken about the dangers of AI, arguing that AI systems need safeguards to prevent them from replacing humans (a topic that he often debated with Google cofounder Larry Page, which eventually led to their falling-out). Musk’s take on AI safety formed the foundation of OpenAI’s goals “to advance digital intelligence in the way that is most likely to benefit humanity as a whole,” per a statement from the company’s website.
Isaac Asimov’s Robot series
Musk found more future AI bait in Isaac Asimov’s series of 37 short stories and six novels centered around, you guessed it, robots. The work grappled with the same topic of AI ethics that Heinlein’s work did. “The tales formulate laws of robotics that are designed to make sure robots do not get out of control,” Isaacson wrote.
To get real dorky with it, one book names one of these rules the Zeroth Law, entailing that “a robot may not harm humanity, or, through inaction, allow humanity to come to harm.” The books had a long-lasting impact; as Isaacson points out, Musk tweeted decades later that “Foundation Series & Zeroth Law are fundamental to the creation of SpaceX.”
This was yet another influence in Musk’s quest with Altman to found OpenAI in a way that benefited humanity. The issue of “AI alignment” aimed to align AGI (artificial general intelligence) with human values and intent—much like how Asimov’s rules in his novels were meant to hinder robots from taking over.
The Hitchhiker’s Guide to the Galaxy by Douglas Adams
Hailing this classic as the novel that most influenced Musk’s “wonder years,” Isaacson wrote that it “helped shape Musk’s philosophy and added a dollop of droll humor to his serious mien.” Rather than making thinly-veiled jokes as Musk tends to do, Adams had more of an ironic take that Musk said helped dig him out of a depressive state.
The story is also centered around understanding the meaning of life. In it, citizens give a supercomputer the task of answering what the universe’s purpose is, which answers with “42” 7 million years later. “The problem, to be quite honest with you, is that you’ve never actually known what the question is,” the computer told the shocked crowd.
Seemingly, Musk brings the lessons from this galaxy—along with his later interest in video games—into every venture that he explores now. “I took from the book that we need to extend the scope of science so that we are better able to ask the questions about the answer, which is the universe,” he said, a thought he has carried with him in creating his own AI venture, xAI, after leaving OpenAI in 2018 when Altman rejected his proposal to run the company.
He said in a July Twitter Spaces Talk that xAI aims “to build a good AGI with the overarching purpose of just trying to understand the universe. The safest way to build an AI is actually to make one that is maximally curious and truth–seeking.”
AI innovations can certainly start to feel supernatural and ring eerily of any famed sci-fi story, as investors poured billions into tech advancements in the field earlier this year. Taking a page from his favored books, Musk told US senators AI was a double-edged sword last week during a summit on the topic and made a plea for regulation on deeper AI.
As one of the early inventors in the field of AI, Musk wants to avoid becoming like the sci-fi trope he once studied, an ego-based scientist who could one day lose control of his invention. Yet he’s since regretted walking away from OpenAI, calling himself “a huge idiot,” for doing so. Perhaps curling up and reading a book can salve the sting.