Is humanity on the brink of an Artificial Intelligence (AI) revolution? Well, you would be forgiven for thinking that as the news has been dominated by that very question. From Bill Gates’ view that robots should be taxed to Stephen Hawking’s concern that artificial intelligence could supplant humans, the news cycle is littered with stories of large strides made in AI.

Our fascination with the rise of AI has been partially fuelled by Hollywood, with films like Total Recall and The Terminator showing a world run by robots and AI. Coupled with this are scientists like Hawking who are concerned that Artificial Intelligence will achieve ‘singularity’. This occurs when an upgradable intelligent device such as Artificial Intelligence (AI) enters a chain of ‘runaway reaction’ of self-improvement cycles. Thereby, causing an new form of ‘intelligent generation’ that appears more and more rapidly. Thus, an intelligence explosion is created, resulting in a powerful superintelligence that would surpass all human intelligence. A process ironically illustrated by Terminator films with Skynet, a powerful superintelligence that takes control of the world.

Yet, this process isn’t confined to Hollywood blockbusters. Since the explosion of the smartphone industry in 2007, the last 10 years has seen cycles of technological innovation, not seen since the Industrial Revolution. We can order devices at the mere command of our voice or ask whether ‘Jon Snow is still alive’. Software like Apple’s Siri and Amazon’s Alexa are becoming such an integral part of our lives, that there are concerns that children are becoming rude as they bark orders at their devices and even the Alexa device has been considered as a witness in a criminal trial.

Artificial Intelligence is not only moving into our homes but in the workplace. Its ability to hoover up data and process patterns within that data, makes the Insurance industry particularly susceptible to AI.

This issue is not purely academic for the industry. The Sunday Times recently reported that Aviva had asked 16,000 employees whether they believed that their jobs could be replaced by automation, offering to retrain them for new roles. Similarly, Japanese insurer, Fukoku Mutual Life Insurance replaced 34 employees with IBM’s Watson Explorer AI. The idea of insurance jobs being automated should not come as a shock. University of Oxford research in 2013, highlighted that insurance agents, appraisers and claims processors as being ‘extremely vulnerable to automation’.

Despite, these negative stories, PWC believes that AI can be transformative for the Insurance sector. In a report published in 2016, it argues that AI is not hype but a new reality that the sector needs to adapt to. For underwriters, this reality will be coming sooner rather than later. The report introduces the concept of ‘Automated and Augmented Underwriting’, the combining of AI and Human Underwriting. What does this new form underwriting look like in practice?

In this new form of underwriting, large classes of underwriting in home, commercial, life and group are automated using sensor data (Internet of Things) and unstructured data (agent/advisor/physician notes) via Bayesian learning or deep learning techniques. The underwriting and new business process is then modelled on soft-robotics and simulation modelling to understand the risk drivers and in turn expand the classes of automated and augmented underwriting.

Finally, the AI systems (NLP and Deep QA system) enlarge the large commercial underwriting and life/disability underwriting to highlight key considerations for the decision makers. The Deep QA system through its deep question answering techniques helps underwriters look for appropriate risk attributes. These attributes help to create the scope for personalised underwriting by a company or individual.

Rather than replacing underwriting, what Artificial Intelligence is doing is improving the analysis and synthesis of the data that is the foundation of the underwriters’ decision-making process. If extended to the Insurers, this has the potential to be revolutionary. Insurers can not only identify emerging risks and analyse them more effectively but they can be in a better positon to determine whether there is an appropriate market for these risks. As a result, new coverage products can be developed in response. Furthermore, the report argues that AI has the capability to generate new sources of revenue from risk and non-risk based information.

So, the question on everybody’s list is can Man and Robot co-exist? It is the view of the PWC report that the solution lies in combining human and machine insights. By placing Man-Machine Learning (MML) at the heart of the organisation, insurers are offered a complementary, value generating capability. In other words, we are combining human originality and capability with the mechanical analysis and synthesis of large data volumes performed by MML.

It is without a shadow of doubt that Artificial Intelligence is going to play a more dominant role in the Insurance industry. What is needed a risk management framework that builds from that premise to help insurers deal with increasing complexity in a connected world. This framework will help maintain that Artificial Intelligence is an innovative force not a disruptive one for Insurers.