IPsoft has launched an artificial intelligence (AI) platform following successful pilots with businesses across the telecoms, finance, energy and media sectors.
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
The system is fronted by an avatar, known as Amelia (pictured). The platform can be used for services such as technology helpdesks, contact centres, procurement processing and to advise field engineers, but can complete many more business processes.
The platform has an understanding of the semantics of language and can learn to solve business process queries like a human. It initially learns using the same manuals as humans – it can read 300 pages in 30 seconds – and then learns through experience and by observing the interactions between human agents and customers.
If it can’t answer a question, it passes the query on to a human but remains in the conversation to learn how to solve similar issues in future.
Amelia understands 20 languages as well as context, and can apply logic and infer implications. Rather than the need for humans to reprogram the system to accomplish new tasks, it automatically creates its own process maps.
Amelia can even understand emotions. By recognising language patters the system can tell if a customer is angry, for example, and change its responses appropriately.
Intelligently automating business processes
IPsoft, which specialises in IT and business process automation, has been working on Amelia since 2001.
Read more on artificial intelligence
Automation software is already reducing the need for labour, but Amelia takes it further by being able to initiate the right process without human involvement.
IPsoft CEO Chetan Dube defined intelligence as the ability to acquire and apply knowledge.
“If a system claims to be intelligent, it must be able to read and understand documents, and answer questions on the basis of that. It must be able to understand processes that it observes. It must be able to solve problems based on the knowledge it has acquired. And when it cannot solve a problem, it must be capable of learning the solution through noticing how a human did it,” he said.
Solving problems with artificial intelligence
In a pilot project with an oil company, Amelia helped engineers fix equipment in remote locations. Because the system has read the manuals, the engineer can ask Amelia to send information about the problem, diagnose the cause and suggest a fix.
A large media company has deployed the IPsoft platform in a central service desk that supports call operators when dealing with complex issues. Within a month it was able to go from solving very few queries independently to 42% of the most common queries. After passing on queries it could not help with, Amelia successfully learned the process from the human agent so it could apply that next time around. By the second month Amelia could answer 64% of those queries independently.