Who is Amelia 2.0?
IPsoft has just announced that their creature, Amelia 2.0 got another step closer to passing the Turing test. But what is Ipsoft? Who is Amelia? What is the Turing test?
All the enthusiastics are surely familiar with these; however for non-expert readers let’s start from the beginning.
Ipsoft is an IT sercives company, established in 1998 by Chetan Dube. It offers solutions to enterprise customers for outsourcing global Internet operations. They focus on four fields of services: Big Data analytics, cloud services, automatic IT services and Amelia. It has a worldwide network of data centres and network operations centres located in North America, Europe and India.
Their project, Amelia is a cognitive knowledge worker, interfaces on human terms. She is a virtual agent who understands what people ask – even what they feel – when they call for service. Using the same instruction manuals as, for example, call centre operators, Amelia can be deployed straight from the cloud in a fraction of the time. She learns as she works and provides high-quality responses consistently, every day of the year, in every language your customers speak.
The first release of Amelia was in 2014, now the newest version has really good result in passing the Turing test.
The Turing test is named after Alan Turing, who was a famous mathematician in the 20th century. He was highly influential in the development of computer science, providing a formalisation of the concepts of algorithm and computation with the Turing machine, which can be considered a model of a general purpose computer.
The Turing test is about to distinguish the machine from another human being. The person, who judges, asks questions through a screen and a keyboard for the two test subjects. None sees and hears each other. One of the test subjects is a human, another is machine, but both of them try to convince the asker that they are humans. It the asker cannot decide after five minutes that which one is human and which one is the machine, then the machine passed the test.
In this October Amelia graduated to version 2.0, bringing the technology another step closer to passing the Turing test, according to IPsoft. Her physical appearance and expressiveness have been transformed to create a more human-like avatar for deeper customer engagement. The developments are that she gained maturity and core understanding capabilities, thereby broadening the range of roles it can assume. Amelia 2.0 has new advances in comprehension and emotional engagement through improvements to memory, contextual comprehension and emotional responsiveness. Her memory functions as the human memory, for instance, she can have more natural conversations. Amelia 2.0 has richer mood and personality attributes, enabling her to personalize the customer service she provides. What is more she has a lot of architectural change, the entire backbone has been rewritten so that the technology can scale seamlessly and remain resilient throughout extreme peaks in volume.
According to IPsoft, she speaks 20 languages and has an emotional quotient. Also new is that the technology can interpret the user’s facial expressions as conveyed by a camera and generate appropriate ones in return.
To sum it up the biggest development in version 2.0 is what is called semantic understanding, however, tools like Amelia are becoming cognitive-like, but they still can’t pass a Turing test.
Other virtual assistant such as Siri and Cortana and Amelia 1.0 did not pass the test. Amelia 2.0’s first performance is going to be on this week at the Gartner Symposium/ITxpo in Orlando. We will see if she passes or not.
I believe that artificial intelligence is not going to beat the human intelligence. I think our feelings and developed emotional intelligence is always going to differentiate ourselves to machines and to animals. Maybe a computer can make better decisions because they decides everything on facts, but there are many situations that our human attributes are needed to make the right decisions.