Alan Turing's Imitation Game has long been a benchmark for machine intelligence. But what it really measures is deception.
Get Started for FREE
Sign up with Facebook Sign up with X
I don't have a Facebook or a X account
Your new post is loading...
Your new post is loading...
|
Aidan Whitehouse's comment,
March 21, 2013 9:15 AM
The possibility of a robot being created that will have the ability to learn multiple different languages without having them programmed into them.
The creators are hoping to be able to use this technology to unlock the ability to create a true artificial neuronal network. If enough robots are built that can truly learn, the ability would be for all of this information to be brought together in the future to create a database with the widest variety of knowledge available in the world, and would allow robots to be created that speak many different languages fluently without each one having to spend months learning the languages.
Rhys Williams's curator insight,
March 22, 2013 12:45 AM
Language Is never static so it is important that AI is able to effectively learn language in an organic manor. |
In 1950, Turing designed a simple test to evaluate whether a computer possessed artificial intelligence comparable to humans; a computer must be able to pass as a human during a series of questions.
Â
Today, Google's text generating deep learning models such as GPT-3 easily pass the Turing test. However, whether these models actually understand their generated output or rather excel at combining human text for specific questions stays up for debate.
Â
This article points out the outdated nature of the Turing test to measure NLP advances which is now evaluated on new benchmarks. The Turing test instead raises ethical concerns for AI and its potential for deceit.Â
Â
It is also interesting to note that NLP models can pass as humans on specific questions but often fail when applied to questions to new domains. Far from resembling human consciousness, current AI remains very specialized and data powered. This motivates the development of new tests to understand model generalization.Â