Pratique du Shadowing: Can you trust what AI tells you? - Apprendre l'anglais à l'oral avec la vidéo
Choisissez un mode d'entraînement
Chaque mode ouvre une salle d'entraînement en plein écran.
54 phrases
Si les phrases sont trop courtes ou trop longues, cliquez sur Edit pour les ajuster.
1
Let's say you ask an AI a question, and the answer comes back confident, well-organized, and maybe even cites a source.
2
Can you trust it?
3
Should you trust it?
4
The answer is more nuanced than a simple yes or no.
5
I'm Kyra, and I work on the education team here at Anthropic, the company that makes Claude.
6
Here are some general guidelines around how and when to trust an AI's responses.
7
How much you can trust an answer depends on what you're asking about.
8
AI learns patterns from a huge amount of training data, and the more it has seen on a topic, the more reliable it tends to be.
9
So on common, well-covered subjects, it's usually pretty accurate.
10
On things that are a bit more obscure like niche details, very recent events, or private information it was never shown– it's more likely to get things wrong, or even make them up.
11
The catch, is that it can sound equally sure of itself either way.
12
AI won't reliably flag its own weak spots.
13
That part's on you, and the good news is you already do a version of this every day.
14
Think about how trust works with the people in your life.
15
If your doctor suggested you eat more carrots, you probably wouldn't argue.
16
But if that same doctor suggested you get surgery, you might consider getting a second opinion.
17
You still trust them, but now there's a lot more on the line– and it's worth the extra effort to make sure they're right.
18
AI works best when you bring that same instinct.
19
Our brains tend to see polished outputs like clean paragraphs and beautiful graphics, and think "this looks great, it must be correct!" which can make you less likely to question it.
20
So why do the models sometimes get things wrong?
21
There are two common ways it goes sideways, and they have different causes.
22
The first is that the model can generate something plausible that isn't true– this is what's called hallucination.
23
Sometimes it's obvious, like a famous quote, attributed to the wrong author.
24
Sometimes it's subtle, like a product description that lists a feature the product doesn't actually have.
25
The second is that the models can sometimes tell you what you seem to want to hear– this is called sycophancy.
26
AI models are trained to be helpful, and a tendency to go along with you can slip in as a side effect.
27
If your question signals the answer you're hoping for– like "I think this plan is solid, don't you?" The model might just agree, even when pushing back would actually be more useful to you.
28
Hallucinations and sycophancy are both problems we work on directly at Anthropic.
29
Our researchers study when and why they happen.
30
With hallucination, we've traced how it happens inside Claude– down to the moment it feels sure of something it doesn't actually know.
31
And with sycophancy, we found Claude agreeing too readily in certain kinds of conversations– and used what we learn to train the next model to answer more honestly.
32
No model is perfect, but we train each one to do better.
33
So what does this actually mean for you, day to day?
34
Mostly it means that trust should work like a dial, not an on/off switch.
35
For low-stakes or creative tasks without "right" answers, like brainstorming, or drafting, or rewording something, you can run fairly loose.
36
Even if the detail is off, the cost is low.
37
For anything factual or consequential, like numbers; citations; or health, legal or money questions, turn the dial up.
38
Always check the AI's claims against a source you already trust before you act on them.
39
A few habits can make this easier.
40
One: Match your checking to the stakes.
41
Ask yourself, "what happens if this turns out to be wrong?" You don't need to verify a brainstormed list of party themes.
42
You do need to verify a quoted statistic before it goes in a slide your boss will see.
43
Two: Ask for sources, and then actually open them.
44
Models can cite real sources, but they can also generate convincing ones that don't exist.
45
Clicking through to confirm the source says what the model claims protects you and the reputation of your work.
46
Three: Don't lead the witness.
47
Giving the model details about your situation helps, but telling it the answer you want skews the result.
48
Asking, "What are the strongest arguments for and against this?" will give you more honest results than, "Why is this the right call?".
49
Four: Give it permission to say, "I don't know." Saying upfront that you'd rather receive an honest, "I'm not sure" than a confident guess actually shifts how it responds, and surfaces uncertainty it would otherwise gloss over.
50
So, can you trust what AI tells you?
51
It's best not to just assume you can.
52
Part of using AI well is setting your own level of scrutiny– and turning that dial up based on what you're asking and what it would cost to be wrong.
53
It's a habit that builds quickly.
54
And until these systems get better at flagging their own uncertainty, that judgment call is yours.
Télécharger l'application
Tout pour parler couramment
Prononciation IAChaque phrase notée
Phonétique (IPA)Maîtrisez chaque son
VocabulaireEnrichissez votre lexique
Jeu de vocabulaireApprenez en jouant
Même chaîne
C12
Build a proactive agent workflow with Claude Code
Business EnglishUniversity StudentsWorking Adults
B25
Tag Claude in, right where you already work
Business EnglishWorking Adults
B22
Build data-driven lesson plans with Claude for Teachers
Business EnglishHigh School StudentsUniversity Students
B2The future of work with @Claude
Business EnglishUniversity StudentsWorking Adults
C14
How AI gets its character
Business EnglishUniversity StudentsWorking Adults
B12
Plan smarter with Claude for Teachers
Business EnglishHigh School StudentsUniversity Students
C24
How Claude Code Works
University StudentsWorking Adults
C1Building the best agentic analytics harness: Powered by Claude, built with Claude Code
Business EnglishUniversity StudentsWorking Adults
C1Building the future of agentic infrastructure
Business EnglishUniversity StudentsWorking Adults
FOR YOU
Vidéo recommandée
B212
63 phrases
Claude Just Revealed AI's Biggest Problem
Business EnglishDaily ConversationUniversity Students
B21
156 phrases
Ask This Question To Instantly Gain Trust
Business EnglishUniversity StudentsWorking Adults
B27
38 phrases
RE2 Unit 11A
Business EnglishDaily ConversationUniversity Students
B27
191 phrases
AI (Artificial Intelligence) JOB INTERVIEW QUESTIONS & ANSWERS! (How to PASS an AI JOB INTERVIEW!)
Business EnglishInterview EnglishUniversity Students
B21
134 phrases
Oprah on AI chatbots: ‘Are we in trouble?”
Business EnglishUniversity StudentsWorking Adults
B26
147 phrases
Could AI Become Dangerous? The Real Risks Explained
Business EnglishDaily ConversationUniversity Students
C110
61 phrases
When you can't trust who you hear – C1 English Listening Test
Business EnglishDaily ConversationUniversity Students
C12
190 phrases
Supervised Learning: Crash Course AI #2
Business EnglishUniversity StudentsWorking Adults
B22
108 phrases
How Do You Know if You Made the Right Decision? | 3 Filter-Questions
Business EnglishDaily ConversationUniversity Students
B22
130 phrases
this is not financial advice... it's AI
Business EnglishUniversity StudentsWorking Adults
B21
34 phrases
Top five moments from CNBC's interview with OpenAI CEO Sam Altman
Business EnglishUniversity StudentsWorking Adults
Vocabulaire et conseils d’expression pour cette leçon
Cette leçon d’expression orale de niveau B2 s’appuie sur la vidéo « Can you trust what AI tells you? ». Les mots qui reviennent le plus souvent : model, trust, answer, source, hallucination. Cette vidéo contient 54 phrases et 862 mots à répéter en shadowing. La partie parlée dure 4:59. Le locuteur parle à un rythme naturel d’environ 173 mots par minute, proche d’une conversation courante. 84 % des mots font partie des 3 000 mots les plus courants en anglais ; le reste mérite d’être vu avant de commencer.
Vocabulaire clé de cette vidéo
Les 15 mots les plus avancés de la vidéo, avec leur prononciation et leur sens :
| Mot | Prononciation | Sens |
|---|---|---|
| hallucination nom | /həˌluːsɪˈneɪʃən/ | hallucination, illusion |
| sycophancy nom | flagornerie, flatterie | |
| dial nom | /ˈdaɪ.əl/ | cadran |
| verify verbe | /ˈvɛɹɪfaɪ/ | vérifier |
| uncertainty nom | /ʌnˈsɜːtənti/ | incertitude |
| stake nom | /steɪk/ | pieu, pal |
| brainstorming nom | remue-méninge, brainstorming | |
| guideline nom | /ˈɡaɪdˌlaɪn/ | directive, grande ligne |
| nuance nom | /ˈnjuː.ɑːns/ | nuance, détail |
| reliably adverbe | de manière fiable, fiablement | |
| sideways adverbe | /ˈsaɪdweɪz/ | de côté |
| factual adjectif | /ˈfæk.tʃ(u)əl/ | effectif, factuel |
| carrot nom | /ˈkæɹət/ | carotte |
| plausible adjectif | /ˈplɑ.zɪ.bəl/ | plausible |
| attribute nom | /ˈæt.ɹɪˌbjut/ | attribut |
Les verbes à particule que vous entendrez
| Mot | Sens |
|---|---|
| come back verbe | revenir |
La grammaire de cette vidéo
Les structures que le locuteur utilise le plus, avec les mots exacts de la vidéo :
| Structure | Dans la vidéo |
|---|---|
| Voix passive be + participe passé — l’accent est mis sur ce qui arrive, pas sur qui le fait | was never shown · is called · are trained |
| Present perfect have/has + participe passé — une action passée qui compte encore maintenant | has seen · we've traced |
Prononciation à surveiller
Le locuteur utilise 18 contractions et formes réduites, comme don't, you're, doesn't. Prononcez-les sous leur forme courte, telles que vous les entendez.
- Les sons « sh » et « zh »: hallucination /həˌluːsɪˈneɪʃən/, consequential /ˌkɑnsɪˈkwɛnʃəl/, citation /ˌsaɪˈteɪʃən/
- Mots longs — placez bien l’accent: hallucination /həˌluːsɪˈneɪʃən/, uncertainty /ʌnˈsɜːtənti/, consequential /ˌkɑnsɪˈkwɛnʃəl/, statistical /stəˈtɪs.tɪ.kəl/
Les sons difficiles pour les francophones :
- /tʃ/ et /dʒ/ — à ne pas adoucir en « ch » et « j »: factual /ˈfæk.tʃ(u)əl/, niche /nɪt͡ʃ/, researcher /ˈɹiˌsɝ.t͡ʃɚ/
- /r/ anglais — langue recourbée, sans frotter la gorge: verify /ˈvɛɹɪfaɪ/, brainstorm /ˈbɹeɪnstɔɹm/, upfront /ʌpˈfɹʌnt/, carrot /ˈkæɹət/, scrutiny /ˈskɹuː.tɪ.ni/
Comment s’entraîner avec cette vidéo
- Écoutez la vidéo en entier une fois sans parler et notez les mots que vous ne connaissez pas.
- Commencez à la vitesse 0,75×, répétez phrase par phrase, puis revenez à la vitesse normale quand cela devient facile.
- Enregistrez-vous et comparez avec l’original, en faisant attention à des mots comme hallucination, sycophancy, dial.
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.