跟讀練習: AI Slop Is Destroying The Internet - 通過影片學習英語口語

選擇練習模式

每種模式都會開啟全屏練習室。

103 句

如果句子過短或過長，請點選 Edit 進行調整。

AI Slop is saturating the internet and things are becoming dramatic pretty quickly.

AI 垃圾內容正在充斥網際網路，情況正迅速變得嚴峻起來。

In an online world where money is made with attention, fake users spread their slop in review sections, generate fake traffic or poison discourse.

在一個靠注意力變現的網路世界裡，虛假使用者會在評論區散播垃圾內容、製造虛假流量，或是擾亂正常討論。

AI has supercharged this and made slop much harder to spot.

人工智慧極大地加劇了這一問題，使得這類粗製濫造的內容更難被識別。

Today about half of internet traffic is bots, the majority of them are used for destructive purposes.

如今，約有一半的網際網路流量來自機器人程式，其中大多數被用於破壞性目的。

It's never been easier to make mediocre content – from the black hole of meaninglessness that is LinkedIN, low effort short videos just engaging enough to hypnotise kids and fry their attention spans, to endless soullessly rewritten books on amazon.

如今製作平庸內容簡直易如反掌——從LinkedIn那毫無意義的黑洞，到那些剛好能讓孩子們沉迷、摧毀他們專注力的低質短影片，再到亞馬遜上無數被改寫得毫無靈魂的書籍。

AI music is invading streaming platforms.

AI音樂正在入侵各大流媒體平臺。

Google AI is summarizing websites instead of sending traffic to them.

谷歌的人工智慧正在對網站內容進行總結，而不再將流量導向這些網站。

On Youtube, new channels publish long form videos multiple times a week with AI generated thumbnails, voices and scripts.

在YouTube上，新頻道每週會多次釋出長影片，這些影片配有AI生成的縮圖、語音和指令碼。

True crime, video essays, science, no space is safe.

真實罪案、影片隨筆、科學領域，沒有哪個領域能倖免。

We’re in the golden era of soulless slop.

我們正處在一個充斥著毫無靈魂的垃圾內容的黃金時代。

And sadly, actual creative human work is used to train these AI models.

令人遺憾的是，這些人工智慧模型的訓練實際上使用了人類的原創作品。

Every reddit comment, original YouTube video or human drawing on deviant art has been sold out to the AI companies.

Reddit 上的每一條評論、YouTube 上的每一個原創影片，還有 DeviantArt 上的每一幅手繪作品，都已經被賣給了人工智慧公司。

Or straight up stolen by them.

或者乾脆就是被他們偷走的。

Without attribution or payment to the actual creators.

未向實際創作者署名或支付報酬。

Creative theft on a scale where it is impossible to protect against it, already putting loads of creatives’ work in danger – so AI companies can get rich.

這種大規模的創意盜竊已經到了無法防範的地步，大量創作者的作品正因此面臨威脅——而這一切只是為了讓AI公司賺得盆滿缽滿。

While this is sad and frustrating, what’s even worse is that generative AI truly has the potential to break the internet irreversibly.

雖然這令人難過和沮喪，但更糟糕的是，生成式AI確實有可能以不可逆轉的方式顛覆網際網路。

By making it harder and harder to tell what is true.

讓人們越來越難以分辨什麼是真相。

At First AI Looked Great!

起初，AI 看起來棒極了！

A kurzgesagt script starts with basic research that is turned into a script and then fact checked in depth by 2-3 people.

Kurzgesagt 的指令碼創作始於基礎研究，研究成果會轉化為指令碼，然後由 2 至 3 人進行深入的事實核查。

We try to confirm our info with what we deem trustworthy, ideally first hand sources, ideally papers.

我們會嘗試通過我們認為可靠的資訊來源來核實我們的資訊，最好是第一手資料，最好是學術論文。

Then we get 1-3 experts for input and critique.

然後我們會邀請1到3位專家提供意見並進行評審。

Fact checking and compiling our sources alone takes around 100 hours per video.

光是事實核查和整理資料來源，每支影片就要花費約100小時。

Of course we make mistakes or oversimplify, and you can disagree with our conclusions – it’s unavoidable, we are only human after all.

當然，我們會犯錯，或者把問題過於簡化，你也完全可以不認同我們的結論——這都是難免的，畢竟我們只是普通人。

But it is fair to say that our process is extensive and after a decade, we know what we are doing.

但公平地說，我們的流程非常完善，經過十年的積累，我們已經駕輕就熟。

When AI appeared we were very excited.

當人工智慧出現時，我們都非常興奮。

A mechanical brain able to super quickly collect information!

一個能夠超快速收集資訊的機械大腦！

So we went to work and it looked amazing!

於是我們開始動手，最終效果看起來棒極了！

And then we started fact checking.

然後我們開始進行事實核查。

We didn’t expect it to be perfect – but it was way worse than we thought.

我們本來就沒指望它能十全十美，但實際情況比我們預想的要糟糕得多。

Confidently Incorrect – AI is SO Bad at This We’ll summarize and generalize our experiences over multiple projects and months condensed into one fake project.

自信滿滿卻漏洞百出——人工智慧在這方面實在太差勁了。我們將把歷時數月、多個專案積累的經驗總結歸納，濃縮成一個虛構的專案案例。

A video about why Brown Dwarfs are the worst and should be ashamed of themselves.

這是一個講解為什麼褐矮星最不靠譜、應該感到羞愧的影片。

So we got all the pro accounts of all the AI models and got to work.

於是我們開通了所有AI模型的專業版賬號，開始著手工作。

Using deep research tools to create a summary and overview of all the information about the failed stars, brown dwarfs.

利用深度研究工具，整理出關於失敗恆星——褐矮星的所有資訊的摘要與概述。

And wow it was great!

哇，太棒了！

Dozens of pages of outlines with unique information nuggets and links to sources!

數十頁大綱，包含獨特的資訊要點和來源連結！

Then we started to look deeper.

隨後我們開始進行更深入的調查。

More than 80% of the info and factoids were pretty solid and we could reconstruct where they were from.

超過80%的資訊和趣聞都相當可靠，我們能夠追溯出它們的來源。

Wikipedia, papers, legit articles.

維基百科、學術論文、正規文章。

What about the rest? It was great stuff, like the speed of brown dwarf superstorms, the nature of their insides or how disappointed their moms are.

那其他內容呢？那些都很精彩，比如褐矮星超級風暴的速度、它們內部的性質，或是它們的“母星”有多失望。

But we could not find where the AI got this info from.

但我們無法找到人工智慧是從哪裡獲得這些資訊的。

Which is not automatically a bad sign – after all, the models contain pretty much all the internet and books ever written.

這本身並不一定是壞事——畢竟，這些模型幾乎囊括了網際網路上的所有內容以及有史以來所有的書籍。

So we asked experts for input – and they flagged the exact same facts, even asking where we found it.

於是我們向專家徵求意見，他們也指出了完全相同的事實，甚至還問我們是從哪裡找到這些資訊的。

To fulfil its goal, to make us happy, the AI had invented or extrapolated information to make brown dwarfs more interesting than they really are.

為了實現讓我們感到愉悅的目標，這臺人工智慧虛構或推導了一些資訊，讓褐矮星顯得比實際情況更有趣。

Like a bad journalist making up details to make a story hit harder or fit a narrative.

就像一個不稱職的記者，為了讓報道更有衝擊力或者符合某種預設說法而編造細節。

Now we wanted to know more and dove in deeper, reading the seemingly more solid sources the AI had given us in full.

現在我們想了解更多，於是深入探究，完整閱讀了人工智慧提供給我們的那些看似更可靠的來源。

One was an article from a news site written by a human journalist – or was it?

其中一篇是來自新聞網站的文章，由人類記者撰寫——但真的是這樣嗎？

It had a very familiar structure and surely purely coincidentally, read like what you get if a human slightly changes AI wording.

它的結構非常眼熟，而且完全是巧合的是，讀起來就像是人稍微修改了AI生成的文字後的樣子。

An AI essay detection tool gave it a 72% match.

一款AI作文檢測工具給出的匹配度為72%。

So an AI article without sources, used as a credible source for AI research.

也就是說，一篇沒有引用來源的人工智慧文章，反倒被當作人工智慧研究的可靠參考資料。

Which makes sense since in 2025 there were already well over 1200 confirmed AI News Websites publishing massive amounts of AI generated misinformation and false narratives.

這是有道理的，因為到2025年，已經有超過1200家經確認的AI新聞網站在釋出大量由AI生成的虛假資訊和不實敘述。

This mix of correct, dubiously sourced and straight up made up information also leads AI to present really shoddy conclusions that sound strong or novel but are often half truths or misrepresentations.

這種由正確資訊、來源可疑的資訊和完全編造的資訊混合而成的內容，還會導致人工智慧給出非常粗糙的結論，這些結論聽起來很有說服力或很新穎，但往往是半真半假或存在誤導的。

Oh well. We salvaged what we could and went back to the drawing board.

好吧，我們儘可能挽回了一些損失，然後重新開始規劃。

And after a few weeks it happened.

幾週後，這件事發生了。

By pure stupid chance, we saw a brown dwarf video popping up, by a relatively new channel.

完全是出於莫名其妙的巧合，我們刷到了一個由一個相對新的頻道釋出的關於褐矮星的影片。

It had hundreds of thousands of views and was pretty good, nice footage, great editing!

它的播放量達到了幾十萬，內容非常棒，畫面拍得很好，剪輯也超讚！

But we had seen the structure and rhythm before.

但我們以前見過這種結構和節奏。

And would you look at that, it contained all the brown dwarf facts we had flagged as unreliable or straight up made up by AI.

你看，裡面包含了我們標記為不可靠或者完全是人工智慧編造的所有關於褐矮星的事實。

This is where the death of the internet begins.

網際網路的消亡正始於此。

Now there is a proper source of brown dwarf misinformation online.

如今網上出現了一個專門傳播關於棕矮星錯誤資訊的渠道。

When the next AI repeats the same research, it will find a transcript from a video with a lot of views.

當下一個人工智慧重複同樣的研究時，它會找到一份來自高播放量影片的文字記錄。

The misinformation is now true.

那些虛假資訊如今反倒成了事實。

It will spread.

它會擴散的。

Even before AI it was pretty hard to find the origin of facts that sound great but are not true – just watch our video about a 100 year old lie.

即便在人工智慧出現之前，要找到那些聽起來很有道理但並不真實的說法的來源就已經相當困難了——不妨看看我們那期關於一個流傳了百年的謊言的影片。

As AI use goes on it may become impossible to know what is true or not.

隨著人工智慧的廣泛應用，我們可能會越來越難以分辨什麼是真相。

The Most Corrosive Lie The problem with AI is how trustworthy it seems. How it is correct enough to seem super smart and how incredibly confidently incorrect it is.

最具腐蝕性的謊言人工智慧的問題在於它看起來是多麼值得信賴。它的正確率高到足以顯得極其聰明，同時它犯下錯誤時又自信得令人難以置信。

Casually lying to your face, often very subtly.

當面隨口撒謊，而且往往非常隱晦。

When you catch it lying, it immediately admits it, vows to never do it again.

當你抓到它撒謊時，它會立刻承認，併發誓再也不這麼做了。

And then it does it again.

然後它又這樣做了一遍。

As eloquent as current language models feel, there is nobody home.

儘管現有的語言模型給人感覺表達能力很強，但它們實際上並沒有真正的意識。

No greater intelligence or consciousness is talking back to you.

並沒有什麼更高等的智慧或意識在回應你。

Current AI is a very complex hammer that does not understand what it is doing or what nails are.

當前的人工智慧就像一把極其複雜的錘子，它根本不明白自己在做什麼，也不知道什麼是釘子。

But we are letting it add new shelves to the library of human knowledge.

但我們正任由它為人類知識的寶庫增添新的書架。

AI is changing quickly, so this may get better, but right now it's pretty grim.

人工智慧發展得很快，所以這種情況可能會有所改善，但目前來說還是相當糟糕的。

Way too many people blindly trust AI.

有太多人盲目信任人工智慧了。

Studies looked into the language of millions of scientific papers published before and after the rise of LLMs.

這些研究分析了大語言模型興起前後發表的數百萬篇科學論文的語言特徵。

They found an abrupt and sharp increase in the frequency of words that AIs like to use.

他們發現，人工智慧常用詞彙的出現頻率出現了突然而急劇的上升。

So it seems clear that now a significant portion of papers have been at least assisted by AI, usually without acknowledgement.

因此，很明顯，現在有相當一部分論文至少得到了人工智慧的協助，而且通常沒有得到說明。

And just in July 2025 it was discovered that a number of researchers had started to sneak hidden messages into their papers.

就在2025年7月，人們發現有不少研究人員開始在他們的論文中偷偷加入隱藏資訊。

In white text or too small for the human eye they prompted AIs to review them positively and not point out flaws.

他們用白色文字或小到肉眼難以辨認的字型，提示人工智慧給出正面評價，不要指出缺陷。

As more and more people are using AI carelessly, the library of human knowledge is getting less and less reliable.

隨著越來越多的人不加謹慎地使用人工智慧，人類的知識儲備正變得越來越不可靠。

Which brings us the last part: How are we using AI?

這就引出了最後一個部分：我們是如何使用人工智慧的？

And will kurzgesagt survive the AI slop age?

那麼Kurzgesagt能在AI生成的垃圾內容時代存活下來嗎？

Will AI Slop Kill Kurzgesagt And Artists?

AI 垃圾內容會扼殺 Kurzgesagt 和藝術家嗎？

On the internet there is only one truly valuable resource: human attention.

網際網路上只有一種真正有價值的資源：人類的注意力。

If current trends continue it is not that far fetched that cheap slop content, stuff just good enough, will soak up the majority of human attention.

如果當前趨勢持續下去，那些廉價的粗製濫造內容、僅僅夠及格線的東西會佔據人類大部分注意力，這並非天方夜譚。

It could make us dumber, less informed, our attention spans even worse, increase political divides and make us neglect real human interaction.

它可能會讓我們變得更愚笨、見聞更淺薄、注意力持續時間更短、加劇政治分歧，並讓我們忽視真實的人際互動。

If AI eats the majority of the attention pie, channels like ours will become unfeasible.

如果人工智慧佔據了大部分注意力份額，我們這類頻道將難以維持運營。

Or forced to downsize or use AI themselves to be able to compete. We don’t want to play this game.

否則他們要麼被迫縮減規模，要麼自己也使用AI才能參與競爭。我們不想玩這種遊戲。

How will we use AI?

我們將如何使用人工智慧？

Like the align tool in adobe illustrator.

就像Adobe Illustrator裡的對齊工具一樣。

If you have a bunch of boxes and you want them to line them up, you can do this manually, one by one.

如果你有一堆箱子，想把它們排成一排，你可以一個一個地手動擺放。

Or you can just select them, click “align” and have them perfectly aligned in an instant.

或者你也可以直接選中它們，點選“對齊”，瞬間就能讓它們完美對齊。

It’s the same with AI programing tools for animation or using it as a faster google alternative.

動畫領域的AI程式設計工具，或是將其用作更高效的谷歌替代工具，情況也同樣如此。

AI is a helpful tool, but the creativity and integrity is still ours.

AI 是一個有用的工具，但創造力和誠信仍然屬於我們。

So dear internet, here is what we are offering to you: kurzgesagt is made by humans, for humans and it will remain that way.

親愛的網友們，在此我們向大家作出承諾：Kurzgesagt 是由人類為人類創作的，並且將永遠保持這一原則。

We will produce well researched content and invest a lot of time and our human creativity into our illustrations and animations.

我們將製作經過充分調研的內容，並投入大量時間和人力創意來打造我們的插圖和動畫。

Put in our creative soul into our work.

我們將創意靈魂傾注到工作中。

We will continue to fact check and discuss our research with human experts, to give you the most trustworthy information that we can.

我們將繼續進行事實核查，並與人類專家討論我們的研究，以儘可能為您提供最值得信賴的資訊。

When we make mistakes, they will be our mistakes.

當我們犯錯時，那些錯誤將由我們自己承擔。

100

We would rather quit than make AI slop.

我們寧願停手也不願製作粗製濫造的AI內容。

101

But to go on we need your support.

但要繼續下去，我們需要您的支援。

102

kurzgesagt is almost 70 full time people and a lot of freelancers on top.

Kurzgesagt 現在有近70名全職員工，此外還有大量自由職業者。

103

This is a lot of salaries, software licences, laptops, rent and coffee.

這包括了大量的薪資、軟體許可證、筆記型電腦、租金和咖啡費用。

下載應用

流利口語所需的一切

AI 發音評分逐句打分

音標練習掌握每個音

詞彙學習建立你的詞庫

詞彙遊戲邊玩邊學

掃碼

Download on theApp Store Get it onGoogle Play

同頻道

FOR YOU

關於本課

在本節課程中，學習者將練習聽力和口語，通過分析與AI相關的網際網路內容，幫助提高英語發音和流利度。這種練習將使您更好地理解如何在數字時代識別資訊的真偽，同時掌握更復雜的英文表達。您將接觸到AI技術影響下的內容創作，瞭解在網際網路資訊氾濫的環境中如何以清晰有效的方式進行交流。

重要詞彙與短語

AI Slop（AI垃圾）
網際網路流量（Internet traffic）
假使用者（Fake users）
創作盜竊（Creative theft）
真實資訊（True information）
資訊來源（Sources）
內容創作（Content creation）

練習技巧

在進行口語練習時，請遵循以下建議，以提高您的英語能力：

利用shadow speech技術，與影片音訊雙向對照練習，注意語音語調。
當影片節奏加快時，嘗試在1.5倍速播放，這樣可以幫助您適應自然語速，提高反應能力。
強調關鍵短語時，可以暫停重放，模仿發音和節奏，對提高提高英語發音非常有幫助。
觀看完影片後，嘗試用自己的話複述內容，這樣可以強化您的理解和表達能力，適合於雅思口語練習。
並針對影片中特定的內容，進行深入的討論或寫作練習，來鞏固您所學的關鍵詞彙。

影片中的語法

說話人最常用的結構，並附上影片中的原話：

結構	影片中的用法
被動語態 be + 過去分詞 — 強調發生了什麼，而不是誰做的	is made · has been sold · is turned
現在完成時 have/has + 過去分詞 — 過去發生但與現在仍有關聯的事	has supercharged · It's never been · has been sold

什麼是跟讀法？

跟讀法 (Shadowing) 是一種有科學依據的語言學習技巧，最初開發用於專業口譯員的培訓，並由多語言者Alexander Arguelles博士普及。這個方法簡單而強大：您在聽英語母語原聲的同時立即大聲重複——就像是一個延遲1-2秒緊跟說話者的影子。與被動聽力或語法練習不同，跟讀法強迫您的大腦和口腔肌肉同時處理並模模擬實的講話模式。研究表明它能顯著提高發音準確性，語調，節奏，連讀，聽力理解和口語流利度——使其成為雅思口語備考和真實英語交流最有效的方法之一。

影子跟讀法: 閱讀完整分步指南 →

跟讀練習: AI Slop Is Destroying The Internet - 通過影片學習英語口語

選擇練習模式

下載應用

Queen Ant Clones Another Species

How Caffeine Works

Could Humans Evolve Flight?

You Will Only Like Half of This Video

Would You Upload Your Mind?

Your Brain on Magic Mushrooms

Why Ebola Keeps Coming Back

Let’s Steal a New Moon

What Happens AFTER Nuclear War?

3 Reasons Why Nuclear Energy Is Awesome! 3/3

What Falling in Love Does to You

We Traveled Back in Time. Now Physicists Are Angry.

YOU Own the Moon. And Mars. And Venus.

What Is Life? Is Death Real?

How Antidepressants Work

Your Brain Peaked at Birth

Your Skeleton Is Electric

We Won’t Explore the Galaxy...

Dozens of Nukes Are Missing. Why? HOW?

Something Travels Faster than Light

Jack Dorsey’s New Dessa AI Voice Sounds Scary Human

AI Governance Explained: Ethical, Transparent & Accountable AI for the Future

It’s not just you

The Limitless AI Lie. The Bubble Is Slowly BURSTING.

China just dropped a new frontier model... and nobody can explain how

The Internet is Worse Than Ever – Now What?

Fastest way to become an AI Engineer in 2026

When AIs act emotional

AI Will End Philosophy, Physicist Claims

The Only SaaS Moats Left in 2026

Teacher to Teacher: Is AI reshaping education for the better?

關於本課

重要詞彙與短語

練習技巧

影片中的語法

什麼是跟讀法？