英特爾新任首席執(zhí)行官帕特·基辛格(Pat Gelsinger)為公司制定了一個(gè)雄心勃勃的綱領(lǐng),要重新奪回這家芯片制造商在行業(yè)內(nèi)首屈一指的地位。
根據(jù)這一規(guī)劃,英特爾將擴(kuò)大其自主設(shè)計(jì)的芯片的產(chǎn)能,并開辟一個(gè)名為“英特爾代工服務(wù)”的新部門,為其他制造商供應(yīng)芯片。英特爾花費(fèi)了數(shù)十年的時(shí)間才成為行業(yè)領(lǐng)導(dǎo)者,但是在過去的幾年中,臺(tái)積電和三星憑借更先進(jìn)的制造技術(shù)迅速崛起,跑到了英特爾前面。
周一,基辛格與美國(guó)總統(tǒng)拜登以及谷歌,福特和通用汽車等其他大公司的首席執(zhí)行官一起,出席了一場(chǎng)白宮峰會(huì),目的是解決困擾汽車行業(yè)、并導(dǎo)致許多科技產(chǎn)品生產(chǎn)延遲的芯片短缺問題。
在這場(chǎng)線上會(huì)議之后,基辛格向《財(cái)富》雜志談及了拜登擬議的向芯片行業(yè)提供500億美元財(cái)政支出的計(jì)劃,作為整個(gè)2萬億美元基礎(chǔ)設(shè)施計(jì)劃的一部分?;粮襁€對(duì)競(jìng)爭(zhēng)對(duì)手英偉達(dá)在周一發(fā)布的新產(chǎn)品作出了回應(yīng)——英偉達(dá)正努力趕超英特爾,搶占對(duì)數(shù)據(jù)行業(yè)至關(guān)重要的芯片市場(chǎng)。周一的發(fā)布會(huì)后,英特爾的股價(jià)下跌了4%,英偉達(dá)的股價(jià)上漲了6%。
為簡(jiǎn)明起見,以下是本次采訪經(jīng)編輯后內(nèi)容。
《財(cái)富》雜志:目前只有大約12%的芯片是在美國(guó)制造的,但拜登總統(tǒng)的計(jì)劃是要讓更多芯片在回到美國(guó)本土制造。為什么這一點(diǎn)如此重要?
基辛格:我們必須讓芯片制造回到美國(guó)本土。美國(guó)的制造量已經(jīng)連續(xù)25年下降了。是時(shí)候開始扭轉(zhuǎn)這一趨勢(shì)了。我有一個(gè)大膽的計(jì)劃,目標(biāo)是讓產(chǎn)量重回25年前的三分之一。很多人都覺得,我們必須要控制相當(dāng)大的一部分產(chǎn)能,才能考慮更長(zhǎng)遠(yuǎn)的計(jì)劃。在芯片這一關(guān)鍵領(lǐng)域,僅僅是讓產(chǎn)量不再下跌還遠(yuǎn)遠(yuǎn)不夠。
我們不僅要讓更多的芯片產(chǎn)自美國(guó),還要將其掌握在美國(guó)公司、美國(guó)工程師和美國(guó)發(fā)明家手中。這一點(diǎn)非常關(guān)鍵。每個(gè)人離開[今天的會(huì)議]之后,都會(huì)有很大的收獲,能向前邁進(jìn)一大步?,F(xiàn)在,我們必須說服國(guó)會(huì)議員,讓他們把錢投入到這個(gè)新生產(chǎn)業(yè)中,因?yàn)樗呀?jīng)準(zhǔn)備就緒了。
即使提出了這一建議,我們也無法立即解決整個(gè)芯片行業(yè)的短缺問題。建立芯片制造廠或者代工廠都需要很長(zhǎng)時(shí)間。短期內(nèi)能做些什么呢?
芯片工廠不是一朝一夕建成的。建一個(gè)廠需要四年時(shí)間。但是在這一過程中,我們可以采取哪些措施來加速這些問題的解決呢?我首先要指出的是,根據(jù)我們發(fā)布的代工計(jì)劃,英特爾的代工服務(wù)將開始向其他人提供現(xiàn)成的[芯片設(shè)計(jì)]。我們不需要從現(xiàn)在開始就干等四年,等待新廠建好。而是可以一邊建廠,一邊開始開放我們的工廠,開始更快地著手,交付他們的一些設(shè)計(jì),并讓供應(yīng)鏈更加靈活。
英偉達(dá)今天發(fā)布了一些對(duì)數(shù)據(jù)產(chǎn)業(yè)至關(guān)重要的新芯片,他們特別關(guān)注的一個(gè)重點(diǎn)是,要加快步伐,把人工智能技術(shù)應(yīng)用到中央處理器(CPU)這種標(biāo)準(zhǔn)芯片中。股市反應(yīng)也很迅速。和他們相比,您的競(jìng)爭(zhēng)力是什么?
我們上周發(fā)布了自己的Ice Lake產(chǎn)品(一種用于服務(wù)器的新型處理器),并得到了非常積極的反響。這種處理器顯示出我們?cè)谕貙捜斯ぶ悄芗夹g(shù)的應(yīng)用上表現(xiàn)卓越。是[英偉達(dá)]在回應(yīng)我們,而不是我們回應(yīng)他們。顯然,這種在CPU中加強(qiáng)人工智能技術(shù)的創(chuàng)新點(diǎn)正是英特爾展示自己行業(yè)領(lǐng)導(dǎo)地位之處。
我們還擁有Habana產(chǎn)品線[英特爾在2019年收購(gòu)的專業(yè)AI芯片制造商],毫無疑問,它為我們的發(fā)展鋪平了道路,我們與亞馬遜云的合作就是一個(gè)很好的例證。我想說的是,很明顯,英特爾正是靠CPU技術(shù)發(fā)家的。我們正在發(fā)展人工智能功能,希望在未來,我們將在這一領(lǐng)域處于攻勢(shì),而非守勢(shì)。(財(cái)富中文網(wǎng))
編譯:陳聰聰
英特爾新任首席執(zhí)行官帕特·基辛格(Pat Gelsinger)為公司制定了一個(gè)雄心勃勃的綱領(lǐng),要重新奪回這家芯片制造商在行業(yè)內(nèi)首屈一指的地位。
根據(jù)這一規(guī)劃,英特爾將擴(kuò)大其自主設(shè)計(jì)的芯片的產(chǎn)能,并開辟一個(gè)名為“英特爾代工服務(wù)”的新部門,為其他制造商供應(yīng)芯片。英特爾花費(fèi)了數(shù)十年的時(shí)間才成為行業(yè)領(lǐng)導(dǎo)者,但是在過去的幾年中,臺(tái)積電和三星憑借更先進(jìn)的制造技術(shù)迅速崛起,跑到了英特爾前面。
周一,基辛格與美國(guó)總統(tǒng)拜登以及谷歌,福特和通用汽車等其他大公司的首席執(zhí)行官一起,出席了一場(chǎng)白宮峰會(huì),目的是解決困擾汽車行業(yè)、并導(dǎo)致許多科技產(chǎn)品生產(chǎn)延遲的芯片短缺問題。
在這場(chǎng)線上會(huì)議之后,基辛格向《財(cái)富》雜志談及了拜登擬議的向芯片行業(yè)提供500億美元財(cái)政支出的計(jì)劃,作為整個(gè)2萬億美元基礎(chǔ)設(shè)施計(jì)劃的一部分。基辛格還對(duì)競(jìng)爭(zhēng)對(duì)手英偉達(dá)在周一發(fā)布的新產(chǎn)品作出了回應(yīng)——英偉達(dá)正努力趕超英特爾,搶占對(duì)數(shù)據(jù)行業(yè)至關(guān)重要的芯片市場(chǎng)。周一的發(fā)布會(huì)后,英特爾的股價(jià)下跌了4%,英偉達(dá)的股價(jià)上漲了6%。
為簡(jiǎn)明起見,以下是本次采訪經(jīng)編輯后內(nèi)容。
《財(cái)富》雜志:目前只有大約12%的芯片是在美國(guó)制造的,但拜登總統(tǒng)的計(jì)劃是要讓更多芯片在回到美國(guó)本土制造。為什么這一點(diǎn)如此重要?
基辛格:我們必須讓芯片制造回到美國(guó)本土。美國(guó)的制造量已經(jīng)連續(xù)25年下降了。是時(shí)候開始扭轉(zhuǎn)這一趨勢(shì)了。我有一個(gè)大膽的計(jì)劃,目標(biāo)是讓產(chǎn)量重回25年前的三分之一。很多人都覺得,我們必須要控制相當(dāng)大的一部分產(chǎn)能,才能考慮更長(zhǎng)遠(yuǎn)的計(jì)劃。在芯片這一關(guān)鍵領(lǐng)域,僅僅是讓產(chǎn)量不再下跌還遠(yuǎn)遠(yuǎn)不夠。
我們不僅要讓更多的芯片產(chǎn)自美國(guó),還要將其掌握在美國(guó)公司、美國(guó)工程師和美國(guó)發(fā)明家手中。這一點(diǎn)非常關(guān)鍵。每個(gè)人離開[今天的會(huì)議]之后,都會(huì)有很大的收獲,能向前邁進(jìn)一大步?,F(xiàn)在,我們必須說服國(guó)會(huì)議員,讓他們把錢投入到這個(gè)新生產(chǎn)業(yè)中,因?yàn)樗呀?jīng)準(zhǔn)備就緒了。
即使提出了這一建議,我們也無法立即解決整個(gè)芯片行業(yè)的短缺問題。建立芯片制造廠或者代工廠都需要很長(zhǎng)時(shí)間。短期內(nèi)能做些什么呢?
芯片工廠不是一朝一夕建成的。建一個(gè)廠需要四年時(shí)間。但是在這一過程中,我們可以采取哪些措施來加速這些問題的解決呢?我首先要指出的是,根據(jù)我們發(fā)布的代工計(jì)劃,英特爾的代工服務(wù)將開始向其他人提供現(xiàn)成的[芯片設(shè)計(jì)]。我們不需要從現(xiàn)在開始就干等四年,等待新廠建好。而是可以一邊建廠,一邊開始開放我們的工廠,開始更快地著手,交付他們的一些設(shè)計(jì),并讓供應(yīng)鏈更加靈活。
英偉達(dá)今天發(fā)布了一些對(duì)數(shù)據(jù)產(chǎn)業(yè)至關(guān)重要的新芯片,他們特別關(guān)注的一個(gè)重點(diǎn)是,要加快步伐,把人工智能技術(shù)應(yīng)用到中央處理器(CPU)這種標(biāo)準(zhǔn)芯片中。股市反應(yīng)也很迅速。和他們相比,您的競(jìng)爭(zhēng)力是什么?
我們上周發(fā)布了自己的Ice Lake產(chǎn)品(一種用于服務(wù)器的新型處理器),并得到了非常積極的反響。這種處理器顯示出我們?cè)谕貙捜斯ぶ悄芗夹g(shù)的應(yīng)用上表現(xiàn)卓越。是[英偉達(dá)]在回應(yīng)我們,而不是我們回應(yīng)他們。顯然,這種在CPU中加強(qiáng)人工智能技術(shù)的創(chuàng)新點(diǎn)正是英特爾展示自己行業(yè)領(lǐng)導(dǎo)地位之處。
我們還擁有Habana產(chǎn)品線[英特爾在2019年收購(gòu)的專業(yè)AI芯片制造商],毫無疑問,它為我們的發(fā)展鋪平了道路,我們與亞馬遜云的合作就是一個(gè)很好的例證。我想說的是,很明顯,英特爾正是靠CPU技術(shù)發(fā)家的。我們正在發(fā)展人工智能功能,希望在未來,我們將在這一領(lǐng)域處于攻勢(shì),而非守勢(shì)。(財(cái)富中文網(wǎng))
編譯:陳聰聰
Intel's new CEO Pat Gelsinger has laid out an aggressive agenda to restore the chipmaker's position on top of the semiconductor industry.
Under the plan, Intel will expand manufacturing capacity for its own designs and also offer to make chips for others at a new unit called Intel Foundry Services. Intel spent decades leading the industry, but for the past few years Taiwan Semiconductor and Samsung have leapt ahead with more advanced manufacturing technologies.
On Monday, Gelsinger attended a White House summit meeting with President Biden and CEOs of other companies including Google, Ford, and General Motors. The goal was to address the chip shortage that has hamstrung the auto industry and caused production delays for many tech gadgets.
After the virtual meeting, Gelsinger spoke to Fortune about Biden's proposed $50 billion chip spending program, part of the president's overall $2 trillion infrastructure plan. Gelsinger also addressed new product announcements on Monday from rival Nvidia, which is seeking to win business away from Intel in the data center chip market. Intel's shares declined 4% and Nvidia's gained 6% on Monday after the announcements.
This interview has been edited for length and clarity.
Fortune: Only about 12% of semiconductors are made in the United States but the President's plan is seeking to get more manufacturing back in the country. Why does that matter?
Gelsinger: We have got to get manufacturing on U.S. soil. We've been on a 25-year decline. Time to start reversing that trend. I laid out a moon-shot objective that we get back to one-third of where we were 25 years ago. I think people rally around that view that we have got to put a big number on the table and go get after it. Just stopping the decline ain't good enough on something this critical.
It isn't just that we need more manufacturing here. We want this to be American companies, American engineers, and American ingenuity. That aspect is so critical. Everybody is coming out of [today's meeting] with a bit more skip in their step. And now we have to run over to Congress and say put money behind this baby, because the industry is ready to go.
Even with this proposal, though, we can't solve the industry-wide chip shortage right away. It takes a long time to build a chip fabrication factory, or fab. Is there anything that can help in the short-term?
You don't build fabs overnight. It takes four years to build a new fab. Inside of that, though, what steps can we take to go faster to get some of these things fixed? One thing I pointed out is, as part of our foundry announcement, the Intel Foundry Services, we will start making that available for existing [chip designs]. I don't need to wait for a new fab built four years from now. I can start opening up our fabs and start stepping in more quickly, start getting some of their designs ported, and giving them more supply chain flexibility.
Nvidia today announced some new chips for the data center market, with a particular focus on speeding up artificial intelligence applications with standard chips known as central processing units, or CPUs. The stock market reacted quickly. What's your competitive position with them?
We announced our Ice Lake [a new microprocessor for servers] last week with an extraordinarily positive response. And in Ice Lake, we have extraordinary expansions in the A.I. capabilities. [Nvidia is] responding to us. It's not us responding to them. Clearly this idea of CPUs that are A.I.-enhanced is the domain where Intel is a dramatic leader.
We also have, with our Habana product line [a specialized A.I. chipmaker Intel bought in 2019], unquestionably laid out a very aggressive path and our cloud partnership with Amazon is a great demonstration of that. So clearly, I'd say the idea of CPUs is Intel's provenance. We're now building A.I. into that and we expect this to be an area where we are on the offense, not the defense going forward.