Share Site Map Contact us Home Page
Home > About AI > FAQs > General FAQs

Newsroom

The Premises

Join the Team

Contact

FAQs
General FAQs
VPoK FAQs
Hal FAQs

Become an Ai Sponsor
 General FAQs
What is artificial intelligence?
What is Ai hoping to achieve?
Where is Ai located?
Haven’t others tried this before?
How is Ai actually creating this intelligent machine?
Why are you pursuing the behavioristic approach?
Who develops Hal's brain?
What is the Learning Machine Challenge?
What is artificial intelligence?
According to conventional wisdom, artificial intelligence is a branch of computer science concerned with making computers think. In the 1950s British scientist Alan Turing derived the common standard judging whether a machine can be considered intelligent. The Turing Test states that a machine can be called intelligent if it fools a human into believing it is human through conversation.
What is Ai hoping to achieve?
Ai is the world's first behavioristically oriented research effort dedicated solely to the creation of true artificial intelligence by following Alan Turing's idea of a "child machine". This technology will enable us to interact with computers in human language. Ai has created a “child machine” that learns to converse through interaction – just like a human child. The child – internally code-named Hal - is currently at the developmental level of a 15 month old baby, and has passed the adapted Turing Test for this stage of development. In ten years Hal will reach the conversational ability of an adult human.
Where is Ai located?
Ai is an international project with a research center near Tel Aviv, Israel. The unique work environment of the Ai Research Center contains offices, living quarters, a swimming pool, an orange grove, and is enhanced with the latest technology. Ai blends the entrepreneurial spirit of the business world with the research process of academia. By providing an atmosphere that encourages creativity and camaraderie, Ai’s team members have an ideal place to pursue their work.
Haven’t others tried this before?
Though other research entities have explored creating artificial intelligence, Ai is the first project to use the Turing Test as a means of judging success. There are many projects and companies that operate in adjacent fields - voice recognition, chat agents, speech synthesis - providing the “mouth and ears” for communication with computers. Ai is creating the “brain” for this communication to take place.
How is Ai actually creating this intelligent machine?
Hal - Ai's child machine - is a technological entity based on a learning algorithm, which is acquiring language through training by a team of cognitive scientists and child development experts. While previous efforts have focused on simple input of data to create a conversational agent (“hard-wiring”), Ai’s is the first to take a behavioristic approach, first creating then educating Hal.
Why are you pursuing the behavioristic approach?
Ai believes language is a skill, which like any other skill, is acquired through training and practice. Conversational capabilities cannot be “injected” into the brain but must undergo a developmental process. An adult’s conversational interaction is based on the set of experiences undergone in childhood, adolescence and up to the present. As humans learn language on a trial-and-error, reward-and-punishment basis, so does Hal.
Who develops Hal's brain?
The development of Hal's brain is an evolutionary process. A variety of machine learning algorithms are competing for the position of Hal's brain. The supply of candidate brains is provided by the Learning Machine Challenge (LMC) community and annual LMC competition. Hal's language learning abilities are constantly enhanced, as his brain evolves.
What is the Learning Machine Challenge?
The Learning Machine Challenge (LMC) initiative is an international collaborative research effort, designed to provide an environment and platform for an evolutionary search for language learning algorithms that can function as Hal's brain, to develop appropriate training strategies for these algorithms, and to promote the field of natural language acquisition using Machine learning by attracting and supporting a community of researchers.
  Printable version