يُعد سؤال «ماذا يجب أن آكل؟» من أكثر الأسئلة التي يطرحها مرضى أمراض التهاب الأمعاء على أطبائهم. إلا أن الإجابة لطالما كانت صعبة، نظرا لقلة الدراسات الواسعة التي تناولت تأثير الأنظمة الغذائية على هذه الأمراض، والتي تشمل التهاب القولون التقرحي .
What is the Turing Test?
Table of Contents
The Turing Test, proposed by Alan Turing in 1950, is a test of a machine’s ability to exhibit intelligent behavior equivalent to, or indistinguishable from, that of a human.
Turing introduced the concept in his 1950 paper “Computing Machinery and Intelligence,” which is widely considered a foundational work in the field of artificial intelligence. The original test, initially called the “Imitation Game,” involved a human evaluator engaging in natural language conversations 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 Turing Test.
Example: In 1966, Joseph Weizenbaum created ELIZA, a natural language processing computer program. While ELIZA could convincingly simulate conversation, it did so through pattern matching and substitution, rather than genuine understanding, and is not considered to have passed the turing Test. it highlighted the difference between simulating intelligence and actually possessing it.
What are the criticisms of the Turing Test?
The Turing Test has faced meaningful criticism for being an insufficient or misleading measure of true intelligence,focusing too heavily on deception and linguistic skill rather than genuine cognitive abilities.
Critics argue that passing the Turing Test requires a machine to *simulate* human conversation, not necessarily to *think* like a human. A machine could potentially pass the test by exploiting loopholes in human judgment, such as feigning ignorance or making grammatical errors, without possessing any real understanding. Furthermore, the test is anthropocentric, meaning it measures intelligence based on human standards, potentially overlooking forms of intelligence that are fundamentally different from our own. The “chinese Room Argument” by John Searle is a prominent philosophical objection.
Example: John Searle’s Chinese room Argument (1980) posits a scenario where a person who doesn’t understand Chinese sits in a room and manipulates Chinese symbols according to a set of rules.To an outside observer, it appears the room ”understands” Chinese, but the person inside has no actual comprehension. Searle argues this demonstrates that a machine can manipulate symbols without possessing understanding, and therefore passing the Turing Test doesn’t equate to intelligence.
What are some alternative tests for AI intelligence?
Several alternative tests and frameworks have been proposed to address the limitations of the Turing Test and provide more comprehensive assessments of AI intelligence.
These alternatives include the Winograd Schema Challenge, the Lovelace Test 2.0, and the Coffee Test. The Winograd Schema Challenge focuses on common-sense reasoning, requiring AI to resolve ambiguous pronouns in sentences. The Lovelace test 2.0 assesses a machine’s ability to create something genuinely novel, going beyond simply following instructions. The Coffee Test, proposed by Hector Levesque, challenges an AI to walk into a coffee shop and order a coffee without any prior programming for that specific scenario.
Example: The Winograd Schema Challenge, initiated in 2016, presents AI systems with pairs of sentences that differ by only one or two words, requiring the system to correctly identify the referent of a pronoun. As a notable example: “The trophy doesn’t fit in the brown suitcase as it is indeed too big.” vs. “The trophy doesn’t fit in the brown suitcase because it is too small.” Correctly answering these requires common-sense knowledge, a challenge for many AI systems.
What is the current status of AI passing the Turing Test?
While no AI has definitively “passed” the Turing Test in a universally accepted manner, some programs have achieved results that blur the lines and spark debate.
In 2014, a chatbot named Eugene Goostman, simulating a 13-year-old Ukrainian boy, reportedly convinced 33% of human judges that it was human during a competition at the royal Society in london. However, this result was controversial, as critics argued the chatbot exploited the judges’ expectations and the limited scope of the conversation. The test conditions and the specific implementation of the chatbot were also questioned.There is no consensus that this constituted a true passing of the Turing Test.
Example: The 2014 event at the Royal Society,organized by the university of Reading,involved 30 human judges and a series of five-minute text-based conversations. Eugene Goostman’s success rate of 33% was below the threshold of 30% that was initially set for passing, but the organizers lowered the threshold during the event, leading to criticism of the validity of the result.The program was developed by Vladimir Veselov and Eugene Demchenko.
