Humanising automation: Understanding the five senses of artificial intelligence
In reality, AI's true potential can only be realised when organisations approach it like a portfolio of technologies, with different attributes, that work together to optimise operational efficiency whilst delivering better service levels and improving the customer and user experience.
This concept of a portfolio of technologies working together can be illustrated further by what I refer to as the five senses of AI. They are: the abilities to communicate, monitor, (store and access) knowledge, analyse and act; and are explained in more detail below.
The first sense of AI is communication. It is arguably the most human quality as the technology engages with the user in real-time: listening to them, reading what they say, and responding, either aloud or in writing. Chatbots are a good example. These types of bots are increasingly transforming staff-run service desks, or managed service call centers, freeing up employees for more complex jobs and customer queries.
At Swedbank, Nina, a virtual assistant, uses Natural Language Understanding (NLU) technology to deliver human-like interactions to customers. She is programmed to be able to answer more than 350 different customer questions. During the first three months, Nina achieved an average of 30,000 conversations per month and an ever improving first-contact resolution rate of 78%.
For many years modern technology has been an important tool to capture data. The Internet of Things (IoT) illustrates the concept of monitoring and has become an accepted part of everyday life.
Standard metering and telemetry have been joined by video, voice, and many other forms of structured and unstructured digital data capture. Constant monitoring can assist organisations in gathering the knowledge that they need to thrive.
Data only becomes knowledge when we can remember it, recall it, and when it’s relevant in answering a question. Managing knowledge has typically been supply driven and involved the storage of historic data on a central data repository. However, AI turns knowledge management on its head. It is only of value where there is demand for that knowledges. The investment to predict future needs and to create new content to supplement historic data is a growing trend.
Just as human minds apply thought to knowledge, AI technologies can be deployed to detect patterns, recognise trends, and apply algorithms to determine appropriate actions or to predict future consequences. What’s more, AI can do this at scale. Deep neural networks (DNNs) can perceive patterns humans would miss because the data sets are simply too large for any one mind to accommodate.
The fifth sense of AI is currently the most widely publicised. It involves using technology to complete activities that are carried out by people. One common example of this is robotic process automation (RPA), which can be used for tasks such as resetting a password, placing a customer order, or processing a supplier invoice.
HM Revenue & Customs (HMRC), the UK tax authority, uses RPA to improve customer services and reduce costs for HMRC. Robotics solutions include dashboards that automatically open relevant case files for contact center advisers to answer customer queries more quickly, and end-to-end processing of first-time employer registrations.
Artificial intelligence is set to enhance and transform the way that enterprises and their employees manage and deliver activities to serve their customers. AI will reach its fullest potential when it is deployed holistically. This means taking time on the design and the implementation plan to understand the exact needs of your business and the impact that it will have on your customers.
Interested in hearing industry leaders discuss subjects like this and sharing their use-cases? Attend the co-located IoT Tech Expo, Blockchain Expo, AI & Big Data Expo and Cyber Security & Cloud Expo World Series with upcoming events in Silicon Valley, London and Amsterdam and explore the future of enterprise technology.
- » A blueprint for open source language automation: How enterprises and developers can benefit
- » Building a foundation for innovation: A guide for the CIO
- » How artificial intelligence will affect the future of networks – and what you need to do about it now
- » The big enterprise security problem posed by small – or at least smaller – businesses
- » Why a risk-based approach to cybersecurity can save time and money