What will a future without secrets look like? | Alessandro Acquisti

譯者: Bert Chen 審譯者: Kuo-Hsien Chiang

I would like to tell you a story

我想告訴各位一則故事

connecting the notorious privacy incident

是著名的亞當和夏娃隱私事件

involving Adam and Eve,

是著名的亞當和夏娃隱私事件

and the remarkable shift in the boundaries

以及這10年來, 在公眾與個人之間分界的重大改變

between public and private which has occurred

以及這10年來, 在公眾與個人之間分界的重大改變

in the past 10 years.

以及這10年來, 在公眾與個人之間分界的重大改變

You know the incident.

你知道這起事件

Adam and Eve one day in the Garden of Eden

有一天亞當和夏娃在伊甸園

realize they are naked.

發現他們都沒穿衣服

They freak out.

他們嚇壞了

And the rest is history.

其餘的部分你們都知道了

Nowadays, Adam and Eve

換做是現在的話, 亞當和夏娃

would probably act differently.

可能會有不同的反應

[@Adam Last nite was a blast! loved dat apple LOL]

(twitter)@亞當 昨晚的表現真是精采! 愛死那顆蘋果了

[@Eve yep.. babe, know what happened to my pants tho?]

(twitter)@夏娃 寶貝你知道我的褲子怎麼了嗎?

We do reveal so much more information

在網路上, 我們都比從前透露出更多關於自己的訊息

about ourselves online than ever before,

在網路上, 我們都比從前透露出更多關於自己的訊息

and so much information about us

而這些有關我們的訊息

is being collected by organizations.

正被許多政府機構收集起來

Now there is much to gain and benefit

現在可以從分析個人資訊或是巨量資料中得到許多利益

from this massive analysis of personal information,

現在可以從分析個人資訊或是巨量資料中得到許多利益

or big data,

現在可以從分析個人資訊或是巨量資料中得到許多利益

but there are also complex tradeoffs that come

但是在捨棄隱私權的同時, 也伴隨著複雜的得失交換

from giving away our privacy.

但是在捨棄隱私權的同時, 也伴隨著複雜的得失交換

And my story is about these tradeoffs.

我要講的故事是有關這些得失交換

We start with an observation which, in my mind,

我們先從觀察開始

has become clearer and clearer in the past few years,

在我看來, 過去幾年中, 這個情況已經變得越來越明確

that any personal information

任何個人資訊

can become sensitive information.

都能變成敏感的訊息

Back in the year 2000, about 100 billion photos

回朔到西元2000年的時候,

were shot worldwide,

全世界上所有人約拍出1000億張照片

but only a minuscule proportion of them

但是只有極少數的照片

were actually uploaded online.

被上傳到網路上

In 2010, only on Facebook, in a single month,

在2010年 光是一個月, Facebook用戶就上傳25億張照片

2.5 billion photos were uploaded,

在2010年時 光是一個月, Facebook用戶就上傳25億張照片

most of them identified.

而多數照片上的人都可以被辨識出來

In the same span of time,

在這段時間裡

computers' ability to recognize people in photos

辨認照片內人物的電腦運算能力

improved by three orders of magnitude.

也加快了1000倍

What happens when you combine

如果將這些技術結合起來

these technologies together:

會發生什麼事?

increasing availability of facial data;

取得了更多的臉部資料

improving facial recognizing ability by computers;

改善了電腦臉部辨識的能力

but also cloud computing,

還有雲端運算

which gives anyone in this theater

這會給與現場任何一個人

the kind of computational power

一種運算能力

which a few years ago was only the domain

而這種運算能力在幾年前只專屬於

of three-letter agencies;

那些政府機構

and ubiquitous computing,

這種普及的運算能力

which allows my phone, which is not a supercomputer,

能讓我的普通手機

to connect to the Internet

連上網際網路

and do there hundreds of thousands

在幾秒之內進行數十萬次的人臉辨識

of face metrics in a few seconds?

在幾秒之內進行數十萬次的人臉辨識

Well, we conjecture that the result

我們推測這些

of this combination of technologies

技術結合的結果

will be a radical change in our very notions

會顛覆我們

of privacy and anonymity.

對於隱私權與匿名性最初的想法

To test that, we did an experiment

為了進行測試 我們做了一項實驗

on Carnegie Mellon University campus.

在卡內基美隆大學的校園裡

We asked students who were walking by

我們找路過的學生

to participate in a study,

來參與這項研究

and we took a shot with a webcam,

我們拿視訊攝影機拍照

and we asked them to fill out a survey on a laptop.

請他們用筆電填寫問卷調查

While they were filling out the survey,

他們在填寫問卷的時候

we uploaded their shot to a cloud-computing cluster,

上傳他們的照片到一個雲端運算群組

and we started using a facial recognizer

使用一個臉部辨識系統

to match that shot to a database

將這組照片拿去與

of some hundreds of thousands of images

一個約有數十萬張圖像的資料庫比對

which we had downloaded from Facebook profiles.

這些圖像是我們從Facebook的個人簡介下載下來的

By the time the subject reached the last page

等到受測者填寫到問卷最後一頁的時候

on the survey, the page had been dynamically updated

畫面會更新成辨識器找出的10張最相符的照片

with the 10 best matching photos

畫面會更新成辨識器找出的10張最相符的照片

which the recognizer had found,

畫面會更新成辨識器找出的10張最相符的照片

and we asked the subjects to indicate

我們要求受測者指出

whether he or she found themselves in the photo.

是否有在這些照片中找到他們自己

Do you see the subject?

你有看到這名受測者嗎?

Well, the computer did, and in fact did so

是的, 電腦有找到

for one out of three subjects.

三人之中就有一人被找到

So essentially, we can start from an anonymous face,

基本上 我們能夠從一張不知名的臉開始,

offline or online, and we can use facial recognition

不管是離線或在線 我們都能利用臉部辨識

to give a name to that anonymous face

讓一張不知名的臉找到它的名字

thanks to social media data.

這都是拜社群媒體資料庫所賜

But a few years back, we did something else.

但是幾年前 我們又做其他的事情

We started from social media data,

我們從社群媒體開始著手-

we combined it statistically with data

我們將它與美國社會安全局的資料做結合

from U.S. government social security,

我們將它與美國社會安全局的資料做結合

and we ended up predicting social security numbers,

我們可以猜出個人的社會安全號碼

which in the United States

這在美國是一項極度敏感的個人資訊

are extremely sensitive information.

這在美國是一項極度敏感的個人資訊

Do you see where I'm going with this?

你知道我在講什麼嗎?

So if you combine the two studies together,

所以如果你們將兩種研究結果加在一起,

then the question becomes,

那這個問題就會變成

can you start from a face and,

你能從一張臉開始

using facial recognition, find a name

利用臉部辨識技術 找到他的名字

and publicly available information

找到這個人的公開資訊

about that name and that person,

找到這個人的公開資訊

and from that publicly available information

再從公開資訊

infer non-publicly available information,

推測出那些更加敏感的非公開資訊

much more sensitive ones

推測出那些更加敏感的非公開資訊

which you link back to the face?

然後你再回想起這張臉嗎?

And the answer is, yes, we can, and we did.

答案是可以的, 而且我們也做到了

Of course, the accuracy keeps getting worse.

當然, 準確度還不是很好

[27% of subjects' first 5 SSN digits identified (with 4 attempts)]

在四次嘗試中, 可以辨識出27%受測者的社會安全號碼前五碼

But in fact, we even decided to develop an iPhone app

但事實上 我們甚至決定做一個 iPhone app

which uses the phone's internal camera

利用手機內建像機

to take a shot of a subject

幫受測者拍一張照片

and then upload it to a cloud

然後上傳至雲端網路

and then do what I just described to you in real time:

接下來馬上就像我對大家描述的一樣

looking for a match, finding public information,

即時找出相符的臉, 找出公開資訊

trying to infer sensitive information,

試著推斷敏感的私人資訊

and then sending back to the phone

然後傳回手機

so that it is overlaid on the face of the subject,

這些資訊會顯示在受測者的臉部照片旁

an example of augmented reality,

這是一個擴增實境的例子

probably a creepy example of augmented reality.

也許是一個會令人毛骨悚然的擴增實境案例

In fact, we didn't develop the app to make it available,

事實上我們並沒有讓這個app上市

just as a proof of concept.

只是做為一種觀念的證明

In fact, take these technologies

事實上, 利用這些科技到極致的時候

and push them to their logical extreme.

事實上, 利用這些科技到極致的時候

Imagine a future in which strangers around you

想像一下未來, 你身旁的陌生人

will look at you through their Google Glasses

能透過Google眼鏡來看你

or, one day, their contact lenses,

或者有一天 隱形眼鏡也能做到同樣的事情

and use seven or eight data points about you

使用七或八個有關於你的資訊

to infer anything else

去推測其他

which may be known about you.

可能與你相關的事

What will this future without secrets look like?

沒有秘密的未來會是什麼樣子?

And should we care?

我們應該關心這個嗎?

We may like to believe

我們可能比較願意去相信

that the future with so much wealth of data

一個有這麼多數據資料的未來

would be a future with no more biases,

會是一個沒有偏差的未來

but in fact, having so much information

但是, 事實上, 擁有這麼多資訊

doesn't mean that we will make decisions

不表示我們能夠做出

which are more objective.

更客觀的決定

In another experiment, we presented to our subjects

在另一個實驗裡 我們把求職者的資訊給受測者看

information about a potential job candidate.

在另一個實驗裡 我們把求職者的資訊給受測者看

We included in this information some references

我們的資料含括

to some funny, absolutely legal,

關於一些有趣, 絕對合法

but perhaps slightly embarrassing information

但也許稍微有點尷尬的訊息

that the subject had posted online.

這些都是受測者張貼在網路上的資訊

Now interestingly, among our subjects,

有趣的是 我們實驗的對象中

some had posted comparable information,

有些人也發表了類似的訊息

and some had not.

但有些人則沒有

Which group do you think

你認為哪一組人

was more likely to judge harshly our subject?

比較可能嚴厲批評我們的受測者?

Paradoxically, it was the group

答案出乎意料的是

who had posted similar information,

那些發表類似訊息的人

an example of moral dissonance.

這也是種道德觀念不一致的例子

Now you may be thinking,

現在你可能正在想

this does not apply to me,

這對我來說沒用

because I have nothing to hide.

因為我沒有什麼要藏的東西

But in fact, privacy is not about

但事實上 隱私不只是

having something negative to hide.

有什麼不好的東西要藏起來而已

Imagine that you are the H.R. director

想像你是某個組織的人事主管

of a certain organization, and you receive résumés,

你收到應徵者寄來的履歷

and you decide to find more information about the candidates.

你決定要找出更多該名應徵者的訊息

Therefore, you Google their names

因此 你就在google上搜尋他們的名字

and in a certain universe,

在特定時空

you find this information.

你可以找到這筆資訊

Or in a parallel universe, you find this information.

或是在平行時空 你找到這筆資訊

Do you think that you would be equally likely

你認為你會同樣的

to call either candidate for an interview?

打電話通知應徵者前來面試嗎?

If you think so, then you are not

如果你這麼認為,

like the U.S. employers who are, in fact,

那你就不像美國雇主

part of our experiment, meaning we did exactly that.

事實上, 他們也在我們的實驗當中

We created Facebook profiles, manipulating traits,

我們創造了一些Facebook個人簡介,

then we started sending out résumés to companies in the U.S.,

然後寄送履歷到美國各家公司

and we detected, we monitored,

然後我們偵查、監控

whether they were searching for our candidates,

看是否他們正在上網搜尋我們的應徵者

and whether they were acting on the information

並且依照這些社群媒體上找到的資訊做事.

they found on social media. And they were.

他們真的這麼作.

Discrimination was happening through social media

透過社群媒體, 對技能相當的應徵者們來說

for equally skilled candidates.

也會發生不公平待遇的事情

Now marketers like us to believe

現在行銷的人想讓我們相信

that all information about us will always

所有關於我們的個人資訊都會

be used in a manner which is in our favor.

用在對我們有利的面向

But think again. Why should that be always the case?

但是再想想, 真的會這樣嗎?

In a movie which came out a few years ago,

幾年前上映的一部電影

"Minority Report," a famous scene

「關鍵報告」裡一個著名的場景

had Tom Cruise walk in a mall

就是湯姆克‧魯斯走進一間賣場

and holographic personalized advertising

有一個個人化的雷射投影廣告

would appear around him.

出現在他旁邊

Now, that movie is set in 2054,

那部電影的時空背景設定於2054年

about 40 years from now,

從現在算起 大約是40年之後

and as exciting as that technology looks,

那種技術看起來很精彩

it already vastly underestimates

它已經大大低估

the amount of information that organizations

各組織能夠匯集起有關你個人的資料量

can gather about you, and how they can use it

與他們是如何運用這些資料

to influence you in a way that you will not even detect.

以某一個你無法查覺的方式, 對你造成影響

So as an example, this is another experiment

還有一個例子 這是另一項實驗

actually we are running, not yet completed.

是我們正在進行中的實驗, 還沒有完成

Imagine that an organization has access

想像一個組織能夠進入

to your list of Facebook friends,

你的Facebook好友清單

and through some kind of algorithm

透過某種運算規則

they can detect the two friends that you like the most.

他們可以偵測到你最喜歡的兩個好友

And then they create, in real time,

然後他們就能即時創造出

a facial composite of these two friends.

由這兩個好友所組成的臉部合成照

Now studies prior to ours have shown that people

在我們之前有研究已經顯示

don't recognize any longer even themselves

人們在看臉部合成照, 連他們自己都認不出來

in facial composites, but they react

人們在看臉部合成照, 連他們自己都認不出來

to those composites in a positive manner.

但是他們對那些合成照有正面評價

So next time you are looking for a certain product,

所以下次你在找某項產品

and there is an ad suggesting you to buy it,

此時有一個建議購買的廣告

it will not be just a standard spokesperson.

廣告將不會是一個固定的代言人

It will be one of your friends,

他很可能是你其中一位朋友

and you will not even know that this is happening.

你甚至不知道這種事正發生在你的生活中

Now the problem is that

現在問題就是

the current policy mechanisms we have

目前政策機制是我們必須

to protect ourselves from the abuses of personal information

保護我們自己免於個人資料遭到濫用

are like bringing a knife to a gunfight.

這就像是以卵擊石

One of these mechanisms is transparency,

其中一項機制就是資訊透明化

telling people what you are going to do with their data.

你必須告訴人們你想拿他們資料做什麼

And in principle, that's a very good thing.

原則上 這是一件非常好的事情

It's necessary, but it is not sufficient.

這是應該的, 但是這麼做還不夠

Transparency can be misdirected.

資訊透明化的方向可能會被誤導

You can tell people what you are going to do,

你可以告訴大家你想做什麼

and then you still nudge them to disclose

然後你促使他人揭露

arbitrary amounts of personal information.

或多或少的個人資訊

So in yet another experiment, this one with students,

在另一項實驗中, 實驗對象是學生

we asked them to provide information

我們要求他們提供個人資訊

about their campus behavior,

關於他們在學校裡做的事

including pretty sensitive questions, such as this one.

包括一些相當敏感的問題 就像這一個

[Have you ever cheated in an exam?]

Now to one group of subjects, we told them,

在考試的時候 你有作弊過嗎?

對其中一組受測者, 我們告訴他們

"Only other students will see your answers."

只有其他的同學會看到你的答案

To another group of subjects, we told them,

對另一組受測者 我們告訴他們

"Students and faculty will see your answers."

所有學生和教職員都會看到你的答案

Transparency. Notification. And sure enough, this worked,

透明化 告知. 這真的有用.

in the sense that the first group of subjects

第一組受測者

were much more likely to disclose than the second.

比第二組受測者更有可能公佈事實

It makes sense, right?

合理吧?

But then we added the misdirection.

但是之後我們加入誤導手段

We repeated the experiment with the same two groups,

我們對相同兩組學生重覆進行實驗

this time adding a delay

這次在不同的時間告訴受測者我們是如何使用他們的資料

between the time we told subjects

這次在不同的時間告訴受測者我們是如何使用他們的資料

how we would use their data

這次在不同的時間告訴受測者我們是如何使用他們的資料

and the time we actually started answering the questions.

現在我們知道了

How long a delay do you think we had to add

你認為我們必須要延遲多久時間

in order to nullify the inhibitory effect

為使這種抑制效應無效

of knowing that faculty would see your answers?

而這種效應就是知道教職員會看見你的答案?

Ten minutes?

10分鐘?

Five minutes?

5分鐘?

One minute?

1分鐘?

How about 15 seconds?

15秒怎樣?

Fifteen seconds were sufficient to have the two groups

15秒就足夠讓兩組人

disclose the same amount of information,

透露出相同資訊量

as if the second group now no longer cares

就好像第二組人現在不再關心教職員會看他們的答案

for faculty reading their answers.

就好像第二組人現在不再關心教職員會看他們的答案

Now I have to admit that this talk so far

現在我必須承認目前為止我說的這些話

may sound exceedingly gloomy,

可能聽起來非常沉悶

but that is not my point.

但我要說的不是這個

In fact, I want to share with you the fact that

事實上 我想與大家分享的是

there are alternatives.

還有替代方案

The way we are doing things now is not the only way

我們現在實驗的方式

they can done, and certainly not the best way

並不是唯一可行的方式

they can be done.

當然也不是最好的辦法

When someone tells you, "People don't care about privacy,"

有人告訴你 「沒人會在乎他的隱私」

consider whether the game has been designed

想想看是否這場遊戲已經遭到設計

and rigged so that they cannot care about privacy,

暗中操作 所以他們才不在意隱私權

and coming to the realization that these manipulations occur

逐漸發現這些操作手段的已經入侵到那些能夠能夠保護你的方法中

is already halfway through the process

逐漸發現這些操作手段的已經入侵到那些能夠能夠保護你的方法中

of being able to protect yourself.

逐漸發現這些操作手段的已經入侵到那些能夠能夠保護你的方法中

When someone tells you that privacy is incompatible

有人告訴你隱私

with the benefits of big data,

與巨量資料所帶來的利益是無法共存的

consider that in the last 20 years,

想想看近20年

researchers have created technologies

研究人員已經研發出數套技術

to allow virtually any electronic transactions

讓幾乎所有電子交易

to take place in a more privacy-preserving manner.

能夠在有更高度的隱私環境下進行

We can browse the Internet anonymously.

我們可以匿名瀏覽網頁

We can send emails that can only be read

傳送唯讀電子郵件

by the intended recipient, not even the NSA.

這些電子郵件僅能由指定的收件者閱讀 就連國家安全局都沒辦法查看

We can have even privacy-preserving data mining.

我們甚至能在隱私受到保護的情況下 進行資料開採

In other words, we can have the benefits of big data

另一方面, 在保護隱私權的同時, 我們仍擁有巨量資料所帶來的好處

while protecting privacy.

另一方面, 在保護隱私權的同時, 我們仍擁有巨量資料所帶來的好處

Of course, these technologies imply a shifting

當然 這些技術也可以看出,

of cost and revenues

在資料持有人與資料提供者之間

between data holders and data subjects,

利益的變化

which is why, perhaps, you don't hear more about them.

這也許是為什麼你沒有聽過太多有關這些技術的事情

Which brings me back to the Garden of Eden.

就讓我將話題轉回伊甸園

There is a second privacy interpretation

有第二種

of the story of the Garden of Eden

對伊甸園故事的隱私解釋

which doesn't have to do with the issue

這與亞當和夏娃

of Adam and Eve feeling naked

覺得全身赤裸

and feeling ashamed.

和感到羞恥沒有關係

You can find echoes of this interpretation

你可以在約翰·密爾頓的《失樂園》裡發現對於這個解釋的迴響

in John Milton's "Paradise Lost."

你可以在約翰·密爾頓的《失樂園》裡發現對於這個解釋的迴響

In the garden, Adam and Eve are materially content.

在伊甸園裡 亞當和夏娃只是物品

They're happy. They are satisfied.

他們很快樂 很滿足

However, they also lack knowledge

然而 他們也缺乏知識

and self-awareness.

和自覺

The moment they eat the aptly named

此刻他們恰好吃下名叫

fruit of knowledge,

「知識」的水果

that's when they discover themselves.

就在那時他們才發現自我

They become aware. They achieve autonomy.

他們開始擁有自覺和自主能力

The price to pay, however, is leaving the garden.

然而 所付出的代價就是必須離開伊甸園

So privacy, in a way, is both the means

所以, 隱私權是自由的意義也是代價

and the price to pay for freedom.

所以, 隱私權是自由的意義也是代價

Again, marketers tell us

市場商人告訴我們

that big data and social media

巨量資料與社群媒體

are not just a paradise of profit for them,

不只對於他們是獲利的天堂

but a Garden of Eden for the rest of us.

對我們其餘的人也是座伊甸園

We get free content.

我們可以得到免費的內容

We get to play Angry Birds. We get targeted apps.

我們可以玩憤怒鳥 使用挑選好的app

But in fact, in a few years, organizations

實際上,在幾年之內

will know so much about us,

政府機構將知道許多關於我們的資訊

they will be able to infer our desires

他們能在我們想到之前推斷我們想做的事情

before we even form them, and perhaps

他們能在我們想到之前推斷我們想做的事情

buy products on our behalf

也許在我們知道我們需要這些東西之前, 就替我們購買產品

before we even know we need them.

也許在我們知道我們需要這些東西之前, 就替我們購買產品

Now there was one English author

現在有一名英國作家

who anticipated this kind of future

考慮到未來可能會發生這種情況

where we would trade away

到時候我們可能會為了過舒適的生活

our autonomy and freedom for comfort.

而賤賣我們的自主能力與自由

Even more so than George Orwell,

其著作比喬治·歐威爾還多

the author is, of course, Aldous Huxley.

這名作家當然就是奧爾德斯·赫胥黎

In "Brave New World," he imagines a society

在《美麗新世界》書中 他想像出一個社會

where technologies that we created

那裡的科技是

originally for freedom

我們為了自由而創造的技術

end up coercing us.

最後我們反被科技奴役

However, in the book, he also offers us a way out

然而 在書中 他也提供我們一個逃離

of that society, similar to the path

那個社會的方式 與那條路很像

that Adam and Eve had to follow to leave the garden.

就是亞當和夏娃離開伊甸園的那條路

In the words of the Savage,

就「野蠻人」這個詞而言

regaining autonomy and freedom is possible,

重新找回自主能力和自由是可行的

although the price to pay is steep.

雖然需要付出的代價實在太高

So I do believe that one of the defining fights

所以我相信我們這個時代的

of our times will be the fight

其中一個決定性的戰鬥將會是

for the control over personal information,

為掌控個人資訊而戰

the fight over whether big data will become a force

不管巨量資料是否將成為一股迎向自由的力量

for freedom,

這場戰鬥終將結束

rather than a force which will hiddenly manipulate us.

而不會成為一股暗中操縱我們的力量-

Right now, many of us

現在 我們當中許多人

do not even know that the fight is going on,

甚至都不知道 戰鬥正在進行

but it is, whether you like it or not.

不管你喜不喜歡 這就是現況

And at the risk of playing the serpent,

冒著玩弄魔鬼的危險

I will tell you that the tools for the fight

我告訴各位, 這場戰爭的工具就在這裡

are here, the awareness of what is going on,

了解現在發生什麼事

and in your hands,

就掌握在你手裡

just a few clicks away.

只要用滑鼠點幾下就行了

Thank you.

謝謝大家

(Applause)

(掌聲)