A simple idea and a profound impact on Artificial Intelligence. Publicado originalmente en Short. Sweet. Valuable..
Alan Turing portrait | Source
I n 1950 Alan Turing created what would become the first benchmark for evaluating a machine’s ability to exhibit intelligent human behavior.
Introduced in the seminal paper: “Computing Machinery and Intelligence”, the test was not only scientific but also a philosophical experiment probing the nature of intelligence and consciousness.
Despite its simplicity, Turing’s proposal generated decades of debate and speculation about the potential and limitations of artificial intelligence.
The Turing test is a method for determining if a computer is capable of thinking like a human being.
Turing test | Source
It’s a kind of imitation game that involves three participants:
- Computer
- Human participant
- Human interrogator
The objective is simple but challenging:
The computer must convince the interrogator that it is a human.
Conducting the test:
- The interrogator converses with the human and the machine, asking questions and generating dialogue.
- Conversations are designed to evaluate how participants exhibit human-like intelligence, such as reasoning, creativity…
- The interrogator must then decide which of the two entities is the machine.
If the interrogator cannot reliably tell the machine from the human, the test is passed.
Eugene — the kid that fooled humanity
In 2014, during a demonstration at University of Reading in London, a computer convinced judges that it was a 13-year-old Ukrainian boy named Eugene Goostman.
The test involved a five-minute keyboard conversation, and evaluators determined whether they were speaking to a machine or a human.
The strategy was that the robot could claim knowledge about various subjects, but also had the excuse of not knowing everything due to its age.
This was key in convincing the judges, even if some answers were not entirely accurate.
Can a computer talk like a human? | TED-Ed