For example, some of us were there in 1996 when IBM's program Deep Blue beat chess grandmaster Garry Kasparov.

例如，1996 年 IBM 的 "深藍 "程序擊敗國際象棋大師加里-卡斯帕羅夫時，我們中的一些人就在現場。

Some of you remember that?

你們中有人還記得嗎？

In 1998, when Google launched its search engine.

1998 年，谷歌推出搜索引擎。

Most of this audience was there in 2011, when Watson the computer won Jeopardy and Siri became a pocket virtual assistant.

2011 年，電腦沃森在 "危險 "遊戲中獲勝，Siri 成為口袋裡的虛擬助手，當時大部分觀眾都在現場。

And we definitely all were there in 2023, when ChatGPT passed the bar exam in the top 10%.

2023 年，當 ChatGPT 以前 10% 的成績通過律師資格考試時，我們肯定都在那裡。

You may have realized that these dates are all milestones associated with AI, artificial intelligence, the ability of machines to learn from experience.

您可能已經意識到，這些日期都是與人工智能（AI）有關的里程碑，人工智能是機器從經驗中學習的能力。

Over that time, AI and machine learning have disrupted just about every industry, including my own, education.

在這段時間裡，人工智能和機器學習幾乎顛覆了所有行業，包括我所在的教育行業。

Teachers, students, and parents are wondering, what is learning when software can answer every question?

教師、學生和家長都想知道，如果軟件能回答所有問題，學習還有什麼意義？

Winston Churchill said, never let a good crisis go to waste.

溫斯頓-丘吉爾說過，永遠不要讓好的危機白白浪費。

Teachers would call this a learning moment.

老師們會把這稱為 "學習時刻"。

I see it as an ideal opportunity for us to reimagine the future of education.

我認為這是我們重新構想未來教育的理想機會。

Unfortunately, my industry's response to change is often flight or fight.

不幸的是，我們行業對變革的反應往往是逃避或抗爭。

When a new disruptor becomes overwhelming, like students with cell phones in the classroom, we're tempted to throw in the towel and say, I'm just going to teach the way I always have.

當新的干擾因素變得難以抵擋時，比如學生在教室裡玩手機，我們就會想放棄，說：我還是按照以前的方式教學吧。

That's flight.

這就是飛行。

Or we try to fight.

或者，我們試圖反抗。

We're told to combat artificial intelligence by using software that promises to detect AI-generated products.

我們被告知要通過使用承諾檢測人工智能生成的產品的軟件來對抗人工智能。

But technology evolves way too fast for any technical solution to keep pace.

但技術發展太快，任何技術解決方案都無法跟上。

Teachers are also advised to monitor their students more closely to prevent high-tech cheating.

此外，還建議教師加強對學生的監控，防止高科技作弊。

Big brother and big sister watching over every student's shoulder every minute is not only impossible, it goes against our mission of building trust, responsibility, independence, and a passion for learning.

大哥哥和大姐姐每時每刻都盯著每個學生，這不僅是不可能的，也違背了我們培養信任、責任感、獨立性和學習熱情的使命。

There are other reasons fight or flight won't work.

戰鬥或逃跑不起作用還有其他原因。

There's flight.

還有飛行。

There's fight.

有戰鬥。

For one, AI is everywhere.

首先，人工智能無處不在。

AI-powered apps are making resources increasingly more accessible, removing barriers like cost and language.

人工智能驅動的應用程序使資源越來越容易獲取，消除了成本和語言等障礙。

And overall, AI has improved efficiency and productivity, which obviously doesn't count the hours we use recording AI-enhanced TikToks.

總體而言，人工智能提高了效率和生產力，這顯然還不算我們使用人工智能增強型 TikToks 錄製的時間。

And AI is growing.

而人工智能正在不斷髮展。

In 2022, another milestone year, Forbes reported 2 million images generated in DALI each day. 145 million Americans, 45% of us, use a voice assistant on a regular basis on over 4 billion devices and the global AI market reached $140 billion, expected to grow by 20% per year.

2022 年是另一個具有里程碑意義的年份，據《福布斯》報道，DALI 每天生成 200 萬張影像。1.45億美國人（佔總人口的45%）在超過40億臺設備上定期使用語音助手，全球人工智能市場規模達到1400億美元，預計每年將增長20%。

So it's pretty clear AI technology is popular, it works, and it's making money, all of which means it's not going anywhere soon.

是以，很顯然，人工智能技術很流行、很有效、很賺錢，這一切都意味著它不會很快消失。

So what's the problem?

那麼問題出在哪裡呢？

Well, we probably shouldn't ignore issues with privacy, intellectual property rights, accuracy, bias, even propaganda, and the difficulty of regulating these technologies.

那麼，我們也許不應該忽視隱私、知識產權、準確性、偏見甚至宣傳等問題，以及監管這些技術的難度。

On top of these problems, as AI gets better at generating all kinds of output, educators and parents are concerned with issues of academic integrity.

除了這些問題之外，隨著人工智能在生成各種產出方面越來越出色，教育工作者和家長也開始關注學術誠信問題。

We're worrying it could cause students to miss important context, amplify prejudices and misconceptions, even bypass the learning process altogether.

我們擔心這會導致學生錯過重要的語境，放大偏見和誤解，甚至完全繞過學習過程。

Now, I was there to experience when many in my profession feared how thesauruses, calculators, video, and spellcheck might ruin our students' abilities to read, write, and do math.

現在，我親身經歷了許多同行擔心辭典、計算器、視頻和拼寫檢查會毀掉學生閱讀、寫作和數學能力的時候。

Today, obviously, we find these in just about every classroom.

如今，很明顯，我們幾乎在每個教室裡都能看到這些東西。

In that spirit, some educators have discovered a third choice to flight or fight, adapt.

本著這種精神，一些教育工作者發現了逃跑或戰鬥之外的第三種選擇--適應。

They're making artificial intelligence their co-teacher by using it to brainstorm lessons, evaluate assignments, even serve as a 24-7 student help desk and fact checker.

他們讓人工智能成為他們的共同老師，利用人工智能來集思廣益、評估作業，甚至充當全天候的學生服務檯和事實核查員。

Others are teaching students about AI technology so they can use its benefits more productively and ethically, understand its limitations, and even consider it as a career.

還有一些人正在向學生傳授有關人工智能技術的知識，以便他們能夠更有效、更合乎道德地利用其優勢，瞭解其侷限性，甚至考慮將其作為職業。

Now, these steps may go a long way towards making AI work in our classrooms, but they don't address the fundamental question of how we measure learning when machines and devices can write essays, generate charts and images, code software, and ace tests.

現在，這些措施可能會大大促進人工智能在課堂上的應用，但它們並不能解決一個根本問題，即當機器和設備可以撰寫論文、生成圖表和影像、編寫軟件代碼並通過考試時，我們如何衡量學習效果。

I suggest that the real threat to our children's education is not machine learning and artificial intelligence.

我認為，對我們孩子教育的真正威脅不是機器學習和人工智能。

It's how we've come to measure and value human learning and human intelligence.

我們就是這樣來衡量和評價人類的學習和智慧的。

A basic definition of learning is the process of gaining understanding through study and experience.

學習的基本定義是通過學習和經驗獲得理解的過程。

Since the 1950s, computer programmers have been developing algorithms that train machines to process information and make decisions that improve with each success or failure.

自 20 世紀 50 年代以來，計算機程序員一直在開發算法，訓練機器處理資訊和做出決策，並在每次成功或失敗中不斷改進。

Now, that should sound familiar because machine learning is modeled on the human brain.

現在，這聽起來應該很熟悉，因為機器學習是以人腦為模型的。

Intelligence has been defined as the ability to acquire and apply that knowledge and skills.

智力被定義為獲取和應用知識與技能的能力。

We say a program is artificially intelligent when it can use the information it's received and learned to answer customized inquiries like, what's the best Bluetooth speaker to buy or write me a heartfelt speech for my friend's wedding.

當一個程序可以利用它所接收和學習到的資訊來回答定製化的詢問，比如：買什麼藍牙揚聲器最好，或者為我朋友的婚禮寫一篇情真意切的演講稿時，我們就可以說它是人工智能程序。

The problem with these traditional definitions of learning and intelligence is that they excel at.

這些關於學習和智力的傳統定義的問題在於，它們擅長於

Machines can beat chess masters, win games of trivia, even predict power failures better than human experts because for these kinds of tasks, they can access and analyze more information far faster than we can.

機器可以打敗國際象棋大師，在瑣事遊戲中獲勝，甚至比人類專家更好地預測電力故障，因為對於這些任務來說，它們可以比我們更快地獲取和分析更多資訊。

These definitions also don't reflect the neurodiversity of our students, each of whom thinks differently and brings different understandings and experiences to our classrooms.

這些定義也沒有反映出我們學生的神經多樣性，他們每個人的思維方式不同，給我們的課堂帶來了不同的理解和體驗。

And when we assign too high a value to the accumulation of knowledge, the memorization of procedures, and the recall of facts, it's no wonder students turn to technology for answers.

而當我們把知識的積累、程序的記憶和事實的回憶看得過重時，難怪學生們會向技術尋求答案。

I propose that we redefine and revalue what we consider intelligence.

我建議我們重新定義和評價我們所認為的智慧。

The good news is the means to that end are probably already in our teacher toolkit.

好消息是，我們的教師工具包裡可能已經有了實現這一目標的手段。

They're just buried in there somewhere under a pile of Scantron forms.

它們就埋在一堆 Scantron 表格下面的某個地方。

Let's design more lessons and ask more questions with these five characteristics.

讓我們利用這五個特點設計更多課程，提出更多問題。

First, emphasize live hands-on experiences.

首先，強調現場親身體驗。

Then ask, what did you notice when this happened?

然後問：發生這種情況時，你注意到了什麼？

How might your actions have affected the outcome?

你的行為會對結果產生什麼影響？

How did your team collaborate on this task?

您的團隊是如何合作完成這項任務的？

Problem solving through discovery, exploration, and reflection is part of what we call active learning, which we know to be more effective than old school models like lectures, worksheets, and book reports.

通過發現、探索和反思來解決問題是我們所說的主動學習的一部分，我們知道這比講課、作業本和讀書報告等舊的學校模式更有效。

Second, ask questions about meaning.

第二，提出有關意義的問題。

Encourage students to identify their beliefs.

鼓勵學生明確自己的信仰。

See the bigger picture, generalize concepts from one area to the other.

縱觀全局，將概念從一個領域歸納到另一個領域。

Make learning personal by asking, what does this information mean to you?

通過詢問 "這些資訊對你有什麼意義？

What might it mean to our community or another community?

它對我們的社區或其他社區意味著什麼？

Why is this relevant to anyone?

這與任何人有什麼關係？

And when children ask us, why do I need to learn this?

當孩子們問我們：我為什麼要學這個？

Adults should have a good answer.

成年人應該有一個很好的答案。

Third, ask students to make connections to the real world, to integrate the pieces of what they know.

第三，要求學生將所學知識與現實世界聯繫起來，將所學知識融會貫通。

Studies show students are more engaged when they understand context.

研究表明，學生在瞭解背景後會更加投入。

For example, how does this relate to last week's lesson or to what you're learning in social studies or art class?

例如，這與上週的課程或您在社會學或藝術課上學習的內容有什麼聯繫？

How do your personal, local, and global perspectives compare?

您的個人、地方和全球視角如何比較？

How might this give you insight on the past, present, and future?

這將如何讓你洞察過去、現在和未來？

Fourth, ask students about their feelings.

第四，詢問學生的感受。

Imagine that.

想象一下

Awareness, empathy, putting yourself in someone else's shoes, and playing nice with others are what we call emotional intelligence.

意識、同理心、設身處地為他人著想以及與他人和睦相處就是我們所說的情商。

As much as we're putting AI into devices that are always listening, maybe even watching, AI still can't read the room.

儘管我們將人工智能應用到設備中，讓它們時刻聆聽，甚至觀看，但人工智能仍然無法讀懂房間。

A one-year-old child, even your cat or dog, can sense your mood better than any computer.

一個一歲大的孩子，甚至是你的貓或狗，都比任何電腦更能感知你的情緒。

Now, we don't normally ask students to flex these four muscles, so things may not go smoothly at first.

現在，我們通常不會要求學生鍛鍊這四塊肌肉，所以一開始可能不會很順利。

But let's keep in mind, struggle is a necessary prerequisite for learning.

但我們要記住，奮鬥是學習的必要前提。

Finally, let's have students direct their knowledge and skills towards novel applications.

最後，讓學生將知識和技能用於新穎的應用。

Ask them to apply their current understandings to a new situation or create a new way of looking at an old problem.

要求他們將現有的理解應用到新的情境中，或創造一種新的方法來看待老問題。

Only humans can think divergently like that, outside the box.

只有人類才能這樣發散思維，跳出條條框框。

Computers are the box.

電腦就是盒子。

Artificial intelligence can only follow the rules they've been given.

人工智能只能遵循它們被賦予的規則。

AI can't experience, generalize, reason, reflect, think abstractly, or understand relevance.

人工智能無法體驗、概括、推理、反思、抽象思維或理解相關性。

And it can't access a student's perceptions, interpretations, and feelings.

而且，它無法獲取學生的認知、解釋和感受。

A good education should do all of these things.

好的教育應該做到這些。

This is how we future-proof our classrooms.

這樣，我們的教室才能面向未來。

Make AI a co-teacher, work with the technology's strengths and weaknesses, but most importantly, let's value human intelligence as more than just computer-like processing.

讓人工智能成為共同的老師，與技術的優缺點合作，但最重要的是，讓我們重視人類的智慧，而不僅僅是計算機般的處理。

Get students to use parts of their brains computers don't have.

讓學生使用計算機沒有的大腦部分。

Ask questions AI can't answer.

提出人工智能無法回答的問題。

Devise problems chatbots can't solve.

設計哈拉機器人無法解決的問題。

We can do that by challenging students with authentic experiences that tap into the incredible capacity, diversity, and creativity of their human minds.

要做到這一點，我們就必須用真實的體驗來挑戰學生，挖掘他們頭腦中令人難以置信的能力、多樣性和創造力。

Now, I can picture the day when I'll be able to say to my future self, imagined here, ironically, by artificial intelligence, that we were all there, and used this learning moment to influence the evolution of education for the better.

現在，我可以想象，有一天我可以對人工智能在這裡想象出的未來的自己說，我們都在那裡，並利用這一學習時刻影響了教育的發展，使之變得更好。

Then, we can all confidently say to our students, we helped you find your there, wherever and whenever it may be.

這樣，我們就可以自信地對我們的學生說，無論何時何地，我們都幫助你們找到了自己的方向。

Thank you.

謝謝。