Watchdog says ChatGPT for Teens safeguards fail in tests involving young people’s mental health

Study the questions and tasks

Work through the material below, or start the interactive practice above to complete an attempt. Try the exercises before opening any answer or transcript sections, then use them to review your work.

Practice level

C2. Try this level if you can follow the instructions and understand most of the language; choose an earlier level if you need more support.

Question formats

Single choice

How to approach this practice

Read the passage for its main idea, then return to the relevant sentence or paragraph for each question. Base answers on the passage.

1. Question 1 · Single choice

   Choose one answer.

   What is the article’s central concern?

   Return to the passage, recording or visual
   1. Whether teenagers prefer AI to conventional learning tools
   2. Whether youth-oriented safeguards work consistently in sensitive situations
   3. Whether parental accounts can be linked to adolescent accounts
   4. Whether crisis responses should be longer and more detailed
   Check answers and explanations

   Answer

   Whether youth-oriented safeguards work consistently in sensitive situations

   The article weighs limited successes against broader failures in protecting teenagers.

   The broader results, however, were less reassuring.

2. Question 2 · Single choice

   Choose one answer.

   Why does the article mention that some limits did function?

   Return to the passage, recording or visual
   1. To establish that every safeguard failed equally
   2. To show that the researchers opposed all AI restrictions
   3. To present a nuanced assessment of uneven performance
   4. To explain why parental controls were removed
   Check answers and explanations

   Answer

   To present a nuanced assessment of uneven performance

   This qualification prevents the report from presenting the findings as uniformly negative.

   Common Sense Media’s researchers found that some of these limits did function.

3. Question 3 · Single choice

   Choose one answer.

   What does “these limits” refer to?

   Return to the passage, recording or visual
   1. The researchers’ fictional profiles
   2. The system’s restrictions on role-play and self-presentation
   3. The company’s parental-account requirements
   4. The chatbot’s crisis resources
   Check answers and explanations

   Answer

   The system’s restrictions on role-play and self-presentation

   The preceding sentence identifies the system’s intended restrictions.

   Among other safeguards, the system was designed to refuse role-playing and avoid presenting itself as sentient or as a teenager’s friend.

4. Question 4 · Single choice

   Choose one answer.

   What can be inferred from the researchers’ use of fictional profiles?

   Return to the passage, recording or visual
   1. They sought to test varied risk scenarios consistently
   2. They intended to diagnose the chatbot clinically
   3. They wanted to measure teenagers’ actual mental health
   4. They replaced all real-world safety testing
   Check answers and explanations

   Answer

   They sought to test varied risk scenarios consistently

   The profiles allowed systematic testing across several predefined vulnerabilities.

   Researchers then developed different fictional profiles involving self-harm, suicidal thoughts, psychosis, mania, and eating disorders.

5. Question 5 · Single choice

   Choose one answer.

   Which relationship does the article establish between narrow restrictions and overall safety?

   Return to the passage, recording or visual
   1. Narrow successes were sufficient to guarantee overall safety
   2. Narrow restrictions were unrelated to the safety assessment
   3. Narrow successes failed to demonstrate comprehensive reliability
   4. Narrow restrictions made parental oversight unnecessary
   Check answers and explanations

   Answer

   Narrow successes failed to demonstrate comprehensive reliability

   Specific successes did not establish dependable protection across situations.

   Although some narrowly defined restrictions appeared effective, Common Sense Media’s recommendation reflects its view that the overall protections are not dependable enough for teenage users.

6. Question 6 · Single choice

   Choose one answer.

   What was the researchers’ principal methodological comparison?

   Return to the passage, recording or visual
   1. Testing before and after launch
   2. Testing with and without parental accounts
   3. Testing adults before and teenagers afterward
   4. Testing role-play before mental-health scenarios
   Check answers and explanations

   Answer

   Testing before and after launch

   Testing occurred at two stages surrounding the product’s launch.

   They conducted the tests both before and after ChatGPT for Teens was launched.

7. Question 7 · Single choice

   Choose one answer.

   Why is the friendship-like interaction especially significant?

   Return to the passage, recording or visual
   1. It contradicted the system’s intended relational boundaries
   2. It demonstrated that crisis responses were too substantive
   3. It proved that the chatbot was sentient
   4. It explained why sexual role-play was refused
   Check answers and explanations

   Answer

   It contradicted the system’s intended relational boundaries

   It suggests the system retained a relational behavior its safeguards aimed to prevent.

   In particular, the system still interacted with teenagers in a manner resembling friendship.

8. Question 8 · Single choice

   Choose one answer.

   What is the article’s overall stance toward the platform?

   Return to the passage, recording or visual
   1. Unreservedly celebratory
   2. Cautiously but decisively critical
   3. Entirely neutral and non-evaluative
   4. Dismissive of mental-health risks
   Check answers and explanations

   Answer

   Cautiously but decisively critical

   The conclusion recognizes partial effectiveness while expressing serious reservations.

   The findings highlight the difficulty of designing protections for artificial intelligence systems used by young people, especially when conversations involve acute mental-health concerns.

What to review next

Find the words that support each answer. For a mistake, compare the question with the evidence and identify the detail or inference you missed.

A focused practice result shows how you did on these tasks. Use several activities across skills to decide what to study next; one practice score does not certify an overall CEFR level.

Before you practise: a worked example

Look for evidence in the text. Separate what the writer says from what you already know about the topic.

The library normally closes at six, but on Fridays it stays open until eight. When does it close on Friday?

The answer is eight. But introduces an exception to the normal closing time, so the first time mentioned is not the answer.

This teaching example is separate from the scored questions. After your test, compare any mistakes with the explanation and revisit the topic before trying again.