Sarah Burnett FBCS shares her industry knowledge and insight in a new book entitled The Autonomous Enterprise - Powered by AI. She talks to Johanna Hamilton AMBCS about innovation, automation and how to normalise AI for the next generation.

‘I have been working in technology since graduating with an MSc in Applied Physics back in the 1980s - I've worked in different capacities in hardware and software design and development, business and project / programme management and industry analysis. But, it's the last eight years of working in intelligent automation - and seeing how companies are innovating with AI and bringing new products and services to market quickly that has inspired me to write the book.’

Tell me about AI and autonomous enterprise

I believe there are possibilities for all with AI. There is much written about it as a general purpose technology that will impact many aspects of our business and personal lives, to the same degree as the personal computer, mobile phone or the web . In business, we have seen a lot of disruption by new players who have taken advantage of AI.

I think of it as an invention that is a tool for invention and that it cannot be pigeonholed as ‘not for us’. In the book I discuss how it can help enterprises get ahead of the competition. There is also the need for speed and efficiency, both of which can be boosted with AI. All of these are contributing to the creation of what I call autonomous enterprises - powered by AI.

No organisation can afford to ignore what is going on.

It's such an exciting time in technology, I believe that a guide to the bigger picture will help organisations navigate their way towards autonomous enterprises of the future. Governments need to understand the path we are on too, so that they can support their industries and build a strong skill base for the future.

How do you see AI changing work?

I believe organisational models in office-based environments will change. I discuss the ‘inside out’ model in my book. I don’t mean ‘inside out’ negatively, but as a different way of working. While the machines will do the bulk of routine transactional work, people will be there to handle more complicated or sensitive requirements. They will also activate automations to get through their own routine work, or to get guidance and advice about what they are doing, such as best answers in contact centres, or insights on real-time situations, for example glitches in logistics.

As such, the machines won’t be there just to process transactions but for personal assistance as well and with analytical insights to help with decision support. People will also design, develop, implement and monitor the automations. They will design and deploy software for new or changed requirements. They will not all be techies but subject matter or process experts.

Where can the robots step in?

I provide many examples in the book, including four real-life named case studies that show good examples of how the technology can be used. In the Royal Berkshire Hospital case study, I wrote about enhanced stroke care following the deployment of an AI powered decision support tool for analysis of CTA scan images, and some process automation.

In another case study, I wrote about Siemens Global Business Services’ (GBS) bionic agent. I discuss how it was developed and deployed in-house to handle service tickets that are generated as a result of customer queries. Siemens GBS gets around 7,000,000 service tickets a year and the software is speeding up and increasing the accuracy of the processing.

Is this a positive change for customer service?

It absolutely is when done well and when customers get a timely and quality response to their queries. The Siemens GBS example is using intelligent document processing. In customer contact services we see a lot of chat-based deployments where the chatbots have not been trained well enough for what they are supposed to do. It means they have a very limited understanding of the context of the service and the intents that they are supposed to be able to handle.

In the book I have provided another case study, the one on Calderdale Council, that shows how you can set the bar high for this kind of deployment with on-going development of the chatbot vocabulary and training to provide more and better services. In the case of the council, it has built in safe-guards for sensitive requirements that get automatically routed to a human agent, for example anything to do with bereavement.

So, is the opportunity people-centric?

It can be, particularly when it comes to improving customer service. Employee engagement is becoming a priority as well in the great resignation. By freeing employees from simple repetitive tasks you give them the opportunity to do more interesting work. In addition, you can make it easier for them to do their work with desktop automation, personal virtual assistants and best answers and guides. It should be about increasing job satisfaction and keeping people engaged on this journey of change.

So how far off are we from having a super intelligence?

A long way off, I'm very glad to say! My book is about narrow AI that's developed to address specific requirements and nothing more. For example, the software that assists physicians to spot stroke damage in the brains of patients can only assess the damage to the front of the brain not the back. Neither can it identify other issues like tumours.

What about AI for bad?

There are reports written by computer scientists about the risks of bad AI. In the wrong hands it can be used maliciously for many purposes, from cybercrime to controlling human lives. That is why I believe we need some kind of regulatory control and the sooner the better. The EU has already started with guidelines and more is to come. It is good to see the focus on AI ethics in the industry as well. It is important to embed ethical practices in every step of AI development and application to ensure that we do not create new problems for mankind.

For you

Be part of something bigger, join the Chartered Institute for IT.

Let’s not forget AI for good. I think there's a lot of AI for good that doesn't get mentioned in the media. AI is already helping with the UN sustainability development goals. For example, it is being used extensively in medicine, not just to read scans better, but in other applications including help with research to spot patterns and join the dots with analytics.

There are AI developments that aim to help the elderly stay in their homes, and live independently for longer. There are also emerging developments in agriculture to optimise the feeding, weeding and watering of crops to protect the environment while increasing yields. That will help with food production to reduce world hunger as the population grows.

Tell us about the innovations of AI?

In my book, I talk a lot about innovation enabled by AI. It can help to generate new ideas, enable disruptive business models and enhance many types of product and services. If you have access to large enough datasets in whatever line of business you’re in, then you can find patterns that will allow you to do things differently. I would say that there is also a place within the innovation cycle for creativity in the usual artistic sense and AI is already contributing to these activities as well.

Experimental research in art, writing and abstract reasoning show how AI can enhance the human capacity to look at the world in a different way and, in doing so, open up new possibilities. In many cases, the greatest opportunities for short-term innovation come from combining AI technology with other existing facilities or processes.

I discuss all these ideas in the book.

Will AI make the difference between start-up mediocrity and success?

Yes, absolutely. I believe today you can assess how things are done in an industry and spot what could be improved. AI can help with the analysis and then help bring a compelling proposition to market fast through automation. It is not just start-ups that can benefit, but established businesses too, particularly as they have a lot of historical data that they can analyse to spot opportunities for doing things better.

Is the UK a world leader in AI?

We have a lot of brains here, and they do a lot of fantastic work. I think we're doing well, but there's more that we could do, for example to help our start-ups grow - it is incredibly difficult to do this in the UK. We need to nurture innovation and entrepreneurship to keep up in this very competitive field.

The more people learn about AI and understand the opportunities that it brings the more innovators we will have and the skills to support the industry. That is why it is important to start teaching children early about AI. We could gamify it, to educate children in school and make them interested in the topic.

Buy the book

Sarah Burnett’s book The Autonomous Enterprise - Powered by AI, with a foreword from Professor Leslie Willcocks, Professor Emeritus at London School of Economics and Political Science, will be available to buy in the BCS Bookshop from 7 February 2022.