---
title: "Does passing Turing test show intelligence"
subject: "Artificial Intelligence"
chapter: 1
key: "semester_6/artificial_intelligence:turing-test-shows-intelligence"
source: https://www.bctengineer.com/q/semester_6/artificial_intelligence/turing-test-shows-intelligence
site: BCT Engineering
---

# Does passing Turing test show intelligence

Artificial Intelligence · Chapter 1: Introduction

- Asked in: 2072 Magh, 7 marks; 2073 Bhadra, 7 marks (asked 2 times)
- Video: [The Turing Test - Computerphile](https://www.youtube.com/watch?v=Qbp3LJvcX38) — Computerphile
- Answer checked by Claude Opus 5.5 (high effort) on 2026-10-14
- Page: https://www.bctengineer.com/q/semester_6/artificial_intelligence/turing-test-shows-intelligence
- Whole chapter: https://www.bctengineer.com/md/semester_6/artificial_intelligence/chapter-1

## Question

If the Turing Test is passed, does this show that computers exhibit intelligence? State your reasons.

## Short answer

Passing the Turing Test shows a machine can behave intelligently indistinguishable from a human in conversation, but does not prove genuine understanding or consciousness. It demonstrates behavioral equivalence and requires NLP, reasoning, and learning, yet Searle's Chinese Room argues syntax manipulation lacks semantics, and programs can fool judges via tricks without real intelligence. The test is limited to text, human-centric, and judge-dependent.

## Answer

**Short answer:** Passing the Turing Test shows that a machine can *behave* intelligently enough to be indistinguishable from a human in conversation. Whether this proves real intelligence (understanding, consciousness) is debatable, so most scholars treat it as sufficient evidence of intelligent *behaviour* but not a proof of genuine understanding.

#### The test
Alan Turing (1950) proposed the Imitation Game. A human interrogator chats through a terminal with two hidden parties, a human and a machine. If the interrogator cannot reliably tell which is the machine, the machine passes.

```text
 Human  <----\
              \     +--------------+
               +--> | Interrogator | asks questions
 Machine <---/      +--------------+
 (hidden)            decides: which is the machine?
```

#### Reasons that passing the test does show intelligence
1. **Behavioural argument:** We judge other people's minds only from their behaviour. If a machine behaves the same, we have no fair ground to deny it intelligence.
2. **Wide capability needed:** To pass, the machine needs natural language processing, knowledge representation, reasoning and learning. These are the core abilities of AI.
3. **Objective:** It avoids vague arguments about the definition of "thinking".

#### Reasons that it does not necessarily show intelligence
1. **Chinese Room argument (Searle):** A person following rules to manipulate Chinese symbols can give correct replies without understanding Chinese. Similarly a program may manipulate symbols (syntax) without understanding (semantics).
2. **Imitation, not intelligence:** The machine only has to *fool* the judge. Programs like ELIZA and Eugene Goostman won over some judges using tricks (changing the subject, pretending to be a child), not real understanding.
3. **Human-centred:** A very intelligent machine that does not behave like a human (or makes no human mistakes) would fail, and an unintelligent one that copies human slips might pass.
4. **Limited scope:** The ordinary test is only text conversation. It does not test vision, perception or physical action (the *Total Turing Test* adds these).
5. **Depends on the judge and the time limit.**

#### Conclusion
Passing the test is strong evidence of human-like intelligent behaviour and a useful practical goal for AI, but it does not by itself prove that the computer thinks or understands.

## Check yourself

The answer's key points as multiple-choice questions, with the correct option and why.

1. In the Imitation Game, when does the machine pass?
   A. When it answers faster than the human
   B. When the interrogator cannot reliably tell which hidden party is the machine
   C. When it solves a maths problem
   D. When the human fails
   Answer: B. The machine passes if the interrogator cannot reliably identify it as the machine. (Key point: Turing Test setup)

2. What does Searle's Chinese Room argument claim?
   A. Machines can never converse
   B. Chinese is hard to translate
   C. Programs manipulating symbols (syntax) may lack understanding (semantics)
   D. Humans always beat computers
   Answer: C. A person following rules can give correct Chinese replies without understanding, so a program may too. (Key point: Chinese Room)

3. Which is a reason FOR treating a passed test as showing intelligence?
   A. Tricks like changing the subject can fool judges
   B. The test covers vision and action
   C. It depends on the judge
   D. We judge other people's minds only by behaviour, so we have no fair ground to deny a machine that behaves the same
   Answer: D. The behavioural argument says we infer minds from behaviour, so a machine behaving the same deserves the same credit. (Key point: Behavioural argument)

4. Why is the ordinary Turing Test limited?
   A. It is only text conversation, so it does not test vision, perception or physical action
   B. It needs too many judges
   C. It tests only mathematics
   D. It requires robots
   Answer: A. The ordinary test is text only; the Total Turing Test adds vision and physical action. (Key point: Limits of the test)

5. What is the answer's overall conclusion?
   A. Passing proves the machine thinks and is conscious
   B. Passing is strong evidence of intelligent behaviour but not proof of genuine understanding
   C. Passing shows nothing at all
   D. Passing proves the machine is a human
   Answer: B. It is strong evidence of human-like behaviour, not proof of real understanding. (Key point: Conclusion)

## Related in Artificial Intelligence

- [AI, Turing test, CSP examples](https://www.bctengineer.com/question.md?key=semester_6%2Fartificial_intelligence%3Aturing-test-pass-csp-examples) — Artificial Intelligence, chapter 1
- [Define AI and the Turing test](https://www.bctengineer.com/question.md?key=semester_6%2Fartificial_intelligence%3Awhat-is-ai-turing-test) — Artificial Intelligence, chapter 1
- [AI, Turing test importance, applications](https://www.bctengineer.com/question.md?key=semester_6%2Fartificial_intelligence%3Aai-turing-test-importance-applications) — Artificial Intelligence, chapter 1

## Related in other subjects

- [Multiplication law; applicant passing test](https://www.bctengineer.com/question.md?key=semester_5%2Fprobability_and_statistics%3Alaw-multiplication-electronics-applicant) — Probability and Statistics, chapter 1
- [Chi-square: intelligence vs economic condition (1072)](https://www.bctengineer.com/question.md?key=semester_5%2Fprobability_and_statistics%3Achi-square-intelligence-economic-1072) — Probability and Statistics, chapter 7, asked 3×
- [Chi-square: intelligence vs economic condition (1650)](https://www.bctengineer.com/question.md?key=semester_5%2Fprobability_and_statistics%3Achi-square-intelligence-economic-1650) — Probability and Statistics, chapter 7, asked 3×
- [Interpret: IQ and intelligence](https://www.bctengineer.com/question.md?key=semester_5%2Fcommunication_english%3Ainterpret-iq-intelligence-reconstructed) — Communication English, chapter 1
