# Turing Test The Turing Test is a measure of machine intelligence proposed by [[Alan Turing]] in his 1950 paper "Computing Machinery and Intelligence." In this test, a human evaluator converses via text with both a human and a machine, without knowing which is which. If the evaluator cannot reliably distinguish the machine from the human, the machine is said to have passed the test—demonstrating intelligent behavior equivalent to a human. Turing proposed this test as a practical alternative to the philosophical question "Can machines think?" By focusing on behavior rather than internal processes, the test sidesteps debates about consciousness and qualia. While influential and widely referenced, the Turing Test has been criticized as neither necessary nor sufficient for intelligence—passing might require deception rather than true understanding, and genuinely intelligent systems might fail for reasons unrelated to intelligence. ## The Imitation Game Turing's original formulation: ``` ┌─────────────────────────────────────────────────────┐ │ Evaluator (C) │ │ (human, asks questions) │ └────────────────────┬────────────────────────────────┘ │ text only ┌───────────┴───────────┐ ▼ ▼ ┌─────────────────┐ ┌─────────────────┐ │ Player A │ │ Player B │ │ (computer) │ │ (human) │ │ tries to seem │ │ responds │ │ human │ │ naturally │ └─────────────────┘ └─────────────────┘ If C cannot reliably tell A from B, A passes. ``` ## Turing's Predictions (1950) > "I believe that in about fifty years' time it will be possible to programme computers... to make them play the imitation game so well that an average interrogator will not have more than 70 percent chance of making the right identification after five minutes of questioning." This prediction was roughly accurate—modern chatbots can fool many humans in brief conversations. ## Criticisms | Criticism | Description | |-----------|-------------| | **Chinese Room** | [[John Searle]]'s argument that passing doesn't prove understanding | | **Anthropocentric** | Tests human-likeness, not general intelligence | | **Deception-based** | Rewards trickery over capability | | **Narrow scope** | Ignores perception, embodiment, creativity | | **Easy to game** | Programs can exploit human biases | ## Notable Attempts | Year | System | Result | |------|--------|--------| | **1966** | ELIZA | Fooled some users with simple pattern matching | | **2014** | Eugene Goostman | Claimed 33% success (controversial) | | **2022** | LaMDA | Google engineer believed it was sentient | | **2023** | GPT-4 | Passes informal Turing tests routinely | ## Variants and Extensions | Variant | Description | |---------|-------------| | **Total Turing Test** | Includes visual and physical tests | | **Loebner Prize** | Annual Turing Test competition | | **Winograd Schema** | Tests commonsense reasoning | | **CAPTCHA** | Reverse Turing Test (prove you're human) | ## Modern Relevance With [[Large Language Models (LLMs)]] like GPT-4: - Systems routinely pass informal Turing Tests - The test no longer distinguishes "intelligent" AI - Focus has shifted to benchmarks testing specific capabilities - Questions about consciousness and understanding remain ## References - Turing, A. (1950). "Computing Machinery and Intelligence" - https://en.wikipedia.org/wiki/Turing_test - https://plato.stanford.edu/entries/turing-test/ ## Related - [[Alan Turing]] - [[Artificial Intelligence (AI)]] - [[Chinese Room Argument]] - [[Large Language Models (LLMs)]] - [[Philosophy of Mind]] - [[ELIZA]]