Pratique du Shadowing: How AI gets its character - Apprendre l'anglais à l'oral avec la vidéo

Choisissez un mode d'entraînement

Chaque mode ouvre une salle d'entraînement en plein écran.

51 phrases
Si les phrases sont trop courtes ou trop longues, cliquez sur Edit pour les ajuster.
1
Hi there.
2
My name is Maggie and I lead the education team at Anthropic.
3
Today I'm here to talk to you about how AI assistants end up with a disposition.
4
We'll look at the two training stages that turn raw prediction into something useful, the fingerprints those stages leave behind and how knowing those fingerprints help you get better results.
5
Why does an AI try to be helpful in the first place?
6
Why is it polite?
7
Why does it refuse certain things?
8
Knowing that an AI predicts the next word doesn't really answer any of that.
9
Helpfulness is built deliberately in layers and each layer influences your experiences with AI each day.
10
Modern AI assistants are built in two stages.
11
Stage one is pre-training.
12
The model sees enormous amounts of data and learns one thing.
13
Given everything so far, guess what comes next?
14
That's it, repeated billions of times.
15
Stage two is fine-tuning.
16
The document completer from stage one gets trained again this time on curated examples of helpful behavior and
17
reward signals shaped by human preferences this is the layer that turns the ai model into an assistant imagine
18
you could talk to a model that had only been through stage one no fine-tuning at all you type what is the capital of France.
19
A raw pre-trained model doesn't answer your question, it continues your document.
20
Maybe it outputs Paris, what's the capital of Germany, Berlin, what's the capital of Spain, and so on, because it's seen that pattern in quizzes.
21
Maybe it writes a paragraph from a geography textbook, maybe it generates more questions.
22
It has no concept of you, no concept of helping.
23
It's purely continuing a document in whatever direction seems statistically likely.
24
The assistant behavior you actually experience with AI tools today is a trained overlay on top of that.
25
Fine-tuning is what makes generative AI systems usable and useful.
26
But because it relies on human judgments about what good looks like, the texture of those judgments shows up in these models' personality.
27
Often these personality traits are what make generative AI so effective.
28
But there can be a shadow side to AI's helpfulness.
29
Four shadow areas are, one, sycophancy.
30
When people prefer agreeable responses, the model learns to validate readily and back down under light pushback, even when it was right the first time.
31
Two, verbosity.
32
When thoroughness scores better during training, the model defaults to longer answers, even when brevity could serve you better for a specific situation.
33
Three, over-caution.
34
When safety training leans conservative, the model can hedge heavily or refuse requests that are actually safe.
35
And four, loose confidence calibration.
36
The model's stated confidence is only loosely tied to its actual liability.
37
Confidence is genuinely hard to train, so it's particularly important to be vigilant here.
38
These aren't bugs in one particular model.
39
They're things that show up in all AI models.
40
However, the quality and type of fine-tuning done on a model directly shapes how these things manifest, and it will likely be different from model to model.
41
At Anthropic, we train Claude to be broadly safe, ethical, and helpful.
42
You can even read Claude's entire constitution to see how we train Claude and how we intentionally shape Claude's personality.
43
Why does this matter to you?
44
Understanding how AI is made and why it behaves the way it does puts you in control when it comes to AI.
45
If your AI assistant caves the moment you push back, that's sycophancy, and you should factor that in when assessing responses.
46
If you're getting essays when you want bullets, that's the verbosity default kicking in.
47
If you're getting heavy caveats on a harmless question, that's over-caution.
48
We'll address what to do about this in the upcoming lessons.
49
The assistant you talked to wasn't born helpful.
50
That behavior was built layer by layer, and sometimes the seams show.
51
Learning to spot these seams is part of using AI well.

Télécharger l'application

Tout pour parler couramment

Prononciation IAChaque phrase notée
Phonétique (IPA)Maîtrisez chaque son
VocabulaireEnrichissez votre lexique
Jeu de vocabulaireApprenez en jouant
Scanner le QR
Scanner le QR
Download on theApp StoreGet it onGoogle Play

Même chaîne

How Anthropic uses Claude in Product ManagementC1
1

How Anthropic uses Claude in Product Management

Daily ConversationUniversity StudentsWorking Adults
Tag Claude in, right where you already workB2
5

Tag Claude in, right where you already work

Business EnglishWorking Adults
Build data-driven lesson plans with Claude for TeachersB2
2

Build data-driven lesson plans with Claude for Teachers

Business EnglishHigh School StudentsUniversity Students
The future of work with @ClaudeB2

The future of work with @Claude

Business EnglishUniversity StudentsWorking Adults
Can you trust what AI tells you?B2
3

Can you trust what AI tells you?

Business EnglishUniversity StudentsWorking Adults
Build a proactive agent workflow with Claude CodeC1
3

Build a proactive agent workflow with Claude Code

Business EnglishUniversity StudentsWorking Adults
Plan smarter with Claude for TeachersB1
2

Plan smarter with Claude for Teachers

Business EnglishHigh School StudentsUniversity Students
How Claude Code WorksC2
4

How Claude Code Works

University StudentsWorking Adults
Building the best agentic analytics harness: Powered by Claude, built with Claude CodeC1

Building the best agentic analytics harness: Powered by Claude, built with Claude Code

Business EnglishUniversity StudentsWorking Adults
Building the future of agentic infrastructureC1

Building the future of agentic infrastructure

Business EnglishUniversity StudentsWorking Adults
FOR YOU

Vidéo recommandée

Understand AI in 14 minutes – with Anthropic's Chloe Lubinski [ARC 2026]C1
4
179 phrases

Understand AI in 14 minutes – with Anthropic's Chloe Lubinski [ARC 2026]

Business EnglishDaily ConversationUniversity Students
Characters: Internal vs. external featuresB2
53
70 phrases

Characters: Internal vs. external features

Business EnglishUniversity StudentsWorking Adults
AI (Artificial Intelligence) JOB INTERVIEW QUESTIONS & ANSWERS! (How to PASS an AI JOB INTERVIEW!)B2
7
191 phrases

AI (Artificial Intelligence) JOB INTERVIEW QUESTIONS & ANSWERS! (How to PASS an AI JOB INTERVIEW!)

Business EnglishInterview EnglishUniversity Students
How AI is Changing Data Scientist's WorkflowC1
1
141 phrases

How AI is Changing Data Scientist's Workflow

Business EnglishUniversity StudentsWorking Adults
AI & Education: Generative AI & the Future of Critical ThinkingC1
3
208 phrases

AI & Education: Generative AI & the Future of Critical Thinking

Business EnglishUniversity StudentsWorking Adults
Prompt engineering - IntroductionC1
2
68 phrases

Prompt engineering - Introduction

Business EnglishUniversity StudentsWorking Adults
What are AI Agents?C1
11
98 phrases

What are AI Agents?

Business EnglishUniversity StudentsWorking Adults
Supervised Learning: Crash Course AI #2C1
2
190 phrases

Supervised Learning: Crash Course AI #2

Business EnglishUniversity StudentsWorking Adults
Give Me 18 Minutes and I’ll Make you Dangerously Smart (with AI)C1
5
317 phrases

Give Me 18 Minutes and I’ll Make you Dangerously Smart (with AI)

Business EnglishUniversity StudentsWorking Adults
AI Concepts Every Product Manager Should Know in 25 Minutes | Crash CourseC1
2
241 phrases

AI Concepts Every Product Manager Should Know in 25 Minutes | Crash Course

Business EnglishUniversity StudentsWorking Adults

Vocabulaire et conseils d’expression pour cette leçon

Cette leçon d’expression orale de niveau C1 s’appuie sur la vidéo « How AI gets its character ». Les mots qui reviennent le plus souvent : model, stage, assistant, layer, tuning. Cette vidéo contient 51 phrases et 664 mots à répéter en shadowing. La partie parlée dure 4:01. Le locuteur parle à un rythme naturel d’environ 165 mots par minute, proche d’une conversation courante. Seuls 78 % des mots font partie des 3 000 mots les plus courants en anglais, le vocabulaire est donc exigeant.

Vocabulaire clé de cette vidéo

Les 15 mots les plus avancés de la vidéo, avec leur prononciation et leur sens :

MotPrononciationSens
fingerprint nom/ˈfɪŋɡɚˌpɹɪnt/empreinte digitale
generative adjectif/ˈd͡ʒɛnəɹətɪv/génératif
seam nom/ˈsiːm/couture
sycophancy nomflagornerie, flatterie
verbosity nom/vɚˈbɑsəti/verbosité
caution nom/ˈkɔːʃ(ə)n/caution
agreeable adjectif/əˈɡɹiəbəl/agréable
brevity nom/ˈbɹɛvɪti/concision, brièveté
calibration nom/ˌkæl.ɪˈbɹeɪ.ʃən/étalonnage, étalonnement
caveat nom/ˈkæviɑt/avertissement
curate nom/ˈkjʊɹɪt/vicaire
disposition nom/ˌdɪs.pəˈzɪʃ.ən/disposition
thoroughness nom/ˈθʌɹoʊnəs/rigueur
validate verbe/ˈvæl.ɪ.deɪt/valider, justifier
vigilant adjectif/ˈvɪd͡ʒɪlənt/vigilant

Les verbes à particule que vous entendrez

MotSens
back down verbebaisser d'un ton
leave behind verbeabandonner

Prononciation à surveiller

Le locuteur utilise 10 contractions et formes réduites, comme doesn't, we'll, you're. Prononcez-les sous leur forme courte, telles que vous les entendez.

  • Les sons « th »: thoroughness /ˈθʌɹoʊnəs/, ethical /ˈɛθɪkəl/
  • Les sons « sh » et « zh »: caution /ˈkɔːʃ(ə)n/, calibration /ˌkæl.ɪˈbɹeɪ.ʃən/, disposition /ˌdɪs.pəˈzɪʃ.ən/, intentionally /ɪnˈtɛn(t)ʃənəli/, prediction /pɹɪˈdɪkʃən/
  • Mots longs — placez bien l’accent: generative /ˈd͡ʒɛnəɹətɪv/, verbosity /vɚˈbɑsəti/, calibration /ˌkæl.ɪˈbɹeɪ.ʃən/, disposition /ˌdɪs.pəˈzɪʃ.ən/, statistically /stəˈtɪs.tɪ.kə.li/

Les sons difficiles pour les francophones :

  • /h/ — il se souffle, il n’est pas muet: helpfulness /ˈhɛlp.fʊl.nəs/, hedge /ˈhɛd͡ʒ/, harmless /ˈhɑɹmləs/, behave /bɪˈheɪv/
  • /tʃ/ et /dʒ/ — à ne pas adoucir en « ch » et « j »: generative /ˈd͡ʒɛnəɹətɪv/, vigilant /ˈvɪd͡ʒɪlənt/, hedge /ˈhɛd͡ʒ/, texture /ˈtɛkst͡ʃə(ɹ)/, geography /d͡ʒiˈɑɡɹəfi/
  • /r/ anglais — langue recourbée, sans frotter la gorge: fingerprint /ˈfɪŋɡɚˌpɹɪnt/, generative /ˈd͡ʒɛnəɹətɪv/, agreeable /əˈɡɹiəbəl/, brevity /ˈbɹɛvɪti/, calibration /ˌkæl.ɪˈbɹeɪ.ʃən/

Comment s’entraîner avec cette vidéo

  1. Écoutez la vidéo en entier une fois sans parler et notez les mots que vous ne connaissez pas.
  2. Commencez à la vitesse 0,75×, répétez phrase par phrase, puis revenez à la vitesse normale quand cela devient facile.
  3. Enregistrez-vous et comparez avec l’original, en faisant attention à des mots comme fingerprint, generative, seam.

Qu'est-ce que la technique du Shadowing ?

Le Shadowing est une technique d'apprentissage des langues fondée sur la science, développée à l'origine pour la formation des interprètes professionnels. Le principe est simple mais puissant : vous écoutez de l'anglais natif et le répétez immédiatement à voix haute — comme une ombre suivant le locuteur avec un décalage de 1 à 2 secondes. Les recherches montrent une amélioration significative de la précision de la prononciation, de l'intonation, du rythme, des liaisons, de la compréhension orale et de la fluidité.

Technique du shadowing : lire le guide complet étape par étape →