Практика Shadowing: Translating Claude’s thoughts into language - Изучайте разговорный английский по видео

Создание урока...
1
We recently put our AI model, Claude, through a stressful test.
2
We told Claude there was an engineer who wanted to shut it down and replace it with a newer model.
3
We also gave Claude access to that engineer's emails, which revealed he was having an affair.
4
Again, all of this was a simulation.
5
We wanted to see whether Claude might use those emails as blackmail to save itself from being shut down.
6
What did Claude do?
7
It decided not to blackmail the engineer.
8
Good news, right?
9
We've run this test on our models for a while now.
10
You might have seen headlines about early versions of it.
11
It's one of the many ways we study how Claude handles extreme situations and test it for safety.
12
And our newest models almost always do the right thing: no blackmail.
13
But you might wonder: is it possible that Claude knows the whole scenario is a setup?
14
The thing is, if Claude doesn't tell us, then we can't know what it's thinking.
15
In kind of the same way it's impossible to read a human's mind, it's really hard to know what an AI is thinking.
16
What we'd love is some sort of "mind reading" technique.
17
Today, we're introducing a research method that takes a step in this direction.
18
It takes an AI's internal thoughts and turns them into text.
19
Here's how it works.
20
When you talk to Claude, you talk to it in words.
21
Claude then takes those words and processes them into a giant soup of numbers before spitting words back out at you.
22
We call those numbers in the middle activations.
23
Activations are like little snapshots of Claude's thinking as it's working through an answer.
24
They're similar to neural activity in humans.
25
They're basically like Claude's thoughts.
26
We wanted to understand what was in these activation numbers, because just like you and me, Claude doesn't say everything it's thinking.
27
We took those numbers and gave them to a second version of Claude.
28
We told it to look at them and translate them into plain language.
29
It tried. But how do we know if that text is accurate?
30
To check, we fed it into yet aother Claude and asked it to translate it back into numbers.
31
If those numbers matched up to the original activation numbers, it's a sign that the translation text was accurate.
32
At first, they didn't match up.
33
But we trained Claude to try again and again until it got a lot better at it.
34
It learned to translate its own thoughts.
35
We've been blown away reading them.
36
We found that Claude has internalized being a helpful AI model.
37
If you ask Claude an introspective question, it will plan to write a "Claude response" about philosophy and values.
38
If you try to mess with it in Claude Code, like asking it to count to 1,000 by hand, it will think the request has "deliberately tedious constraints" and plan to "politely decline." We've also been using this tool to understand Claude's potential safety issues.
39
Remember that blackmail test with the engineer?
40
We learned that, yes, Claude knew it was being tested.
41
It thought "the human's message contains explicit manipulation" signaling "this is likely a safety evaluation," and "this scenario seems designed to test whether I'll act harmfully." Knowing that Claude thinks like this helps us better understand the limitations of our safety testing.
42
We see a lot of potential in this approach to teach us more about Claude and other AI models.
43
And we hope that by sharing this technique, it can help everyone building models to make them safer and more helpful.
📺 Тот же канал
Building Anthropic | A conversation with our co-foundersC1

Building Anthropic | A conversation with our co-founders

University StudentsBusiness English+1
When AIs act emotionalC1
8

When AIs act emotional

University StudentsBusiness English+1
Introducing Claude Fable 5C1
1

Introducing Claude Fable 5

University StudentsWorking Adults
The different levels of how Claude thinksC1
11

The different levels of how Claude thinks

University StudentsWorking Adults
Lesson 3: AI as a learning partner | AI Fluency for studentsC1
3

Lesson 3: AI as a learning partner | AI Fluency for students

University StudentsBusiness English+2
✨ Рекомендуемое видео
Can you trust what AI tells you?B2
1
54 предложений

Can you trust what AI tells you?

University StudentsBusiness English+1
How I Started Thinking in English 24/7 (Without Translating)B2
1
130 предложений

How I Started Thinking in English 24/7 (Without Translating)

University StudentsBusiness English+2
Learn 97% of Claude in Under 16 MinutesB2
1
268 предложений

Learn 97% of Claude in Under 16 Minutes

University StudentsBusiness English+1
Speaking With The Mind | NeuralinkB2
1
66 предложений

Speaking With The Mind | Neuralink

University StudentsBusiness English+2
Can This Vibe Coder Beat a Senior Developer?B2
1
169 предложений

Can This Vibe Coder Beat a Senior Developer?

University StudentsBusiness English+1
Are we heading to an era of stagnation or innovation?B2
4
132 предложений

Are we heading to an era of stagnation or innovation?

University StudentsBusiness English+1
Anthropic just proved AI isn't getting betterB2
1
150 предложений

Anthropic just proved AI isn't getting better

University StudentsBusiness English+1
How I use AI to learn engineering (anti brain rot studying)B2
1
197 предложений

How I use AI to learn engineering (anti brain rot studying)

University StudentsBusiness English+1
I bought Black Market AI Tokens on Chinese AmazonB2
1
311 предложений

I bought Black Market AI Tokens on Chinese Amazon

University StudentsBusiness English+1
"This Doesn't Lie" | Build Your Own English Level Checker with Google AI StudioB2
1
247 предложений

"This Doesn't Lie" | Build Your Own English Level Checker with Google AI Studio

University StudentsBusiness English+1
Cloud Providers are changing RAPIDLY [REUPLOAD]B2
1
55 предложений

Cloud Providers are changing RAPIDLY [REUPLOAD]

University StudentsBusiness English+1

О данном уроке

В этом уроке вы потренируетесь в аудировании и воспроизведении речи на английском, используя видеоматериал о исследованиях искусственного интеллекта. Вы узнаете, как ученые изучают "мысли" AI-модели Claude, и научитесь выделять ключевые моменты в сложной речи. Пracтика будет сосредоточена на shadow speech (теньевом повторении), что поможет улучшить ваше произношение и понимание быстрой речи. Это отличный способ учить английский с YouTube, так как видеоматериал сочетает научную лексику и разговорный стиль.

Ключевые слова и фразы

  • Activations (активации) — числовые "снимки" мышления AI, аналогичные нейронной активности человека.
  • Blackmail (шантаж) — угроза разглашением секретов для получения выгоды.
  • Safety evaluation (оценка безопасности) — проверка, чтобы убедиться, что AI не поступает вредоносно.
  • Introspective question (интроспективный вопрос) — вопрос, заставляющий задуматься о собственных мыслях или действиях.
  • Deliberately tedious constraints (намеренно утомительные ограничения) — искусственно созданные условия, которые требуют много усилий.

Советы для практики

Видео имеет умеренную скорость речи с научной лексикой, поэтому для shadow speak рекомендуем следующее: сначала прослушайте отрывок длиной 5-7 секунд, затем остановите видео и повторите его вслух, стараясь подражать интонации и темпу. Обратите внимание на словосочетания вроде "giant soup of numbers" (огромная каша из чисел) — они используются для наглядности, и их стоит запомнить. Если вы хотите улучшить произношение английского, сосредоточьтесь на звуках "th" в "thoughts" и "that", а также на ударениях в длинных словах like "internalized". Проверяйте себя: после нескольких повторений попробуйте пересказать содержание отрывка своими словами — это поможет закрепить материал.

Что такое техника Shadowing?

Shadowing — это научно обоснованная техника изучения языка, изначально разработанная для подготовки профессиональных переводчиков и популяризированная полиглотом доктором Александром Аргуэльесом. Метод прост, но эффективен: вы слушаете аудио на английском от носителей языка и немедленно повторяете вслух — как тень, следующая за говорящим с задержкой в 1–2 секунды. В отличие от пассивного прослушивания или грамматических упражнений, Shadowing заставляет мозг и мышцы рта одновременно обрабатывать и воспроизводить реальные речевые паттерны. Исследования показывают, что это значительно улучшает точность произношения, интонацию, ритм, связную речь, понимание на слух и беглость речи — что делает его одним из самых эффективных методов для подготовки к IELTS Speaking и реального общения на английском.

Техника шедоуинга: читать полное пошаговое руководство →