As AI continues to evolve, it will arguably and increasingly take over routine decision-making, rule-based processes, and repetitive tasks across many industries, including insurance, reshaping the workforce and shifting the focus from task execution to managing and leveraging AI systems.  Indeed the insurance sector was one of the first to adopt AI for risk profiles.  

Business in general is still struggling with a perception issue around AI, which is often talked about as something that does everything automatically and intelligently.  In reality, we should be thinking of AI simply as another engine or a tool – a technology, not a solution.  Even the idea of AI itself – most people associate it with GenAI and ChatGpt, yet AI has multiple branches, including machine learning, computer vision, pattern recognition and more. Some organisations struggle with this broad concept and submit to the temptation of thinking of it as an application or a strategy in itself; however, the technology is still a long way from becoming an entity that can lead, implement and operate itself to a purposeful end but it will increasingly power applications overlaid by strategic, human-led frameworks.

Effective AI integration

Effective AI integration needs a structured approach in the same way that any other effective IT integration does. People are still in control, but as with any rapid advance in any other chain, improvements in productivity must be weighed against increased margin for error and this risk increases when those making decisions to implement don’t fully understand the strengths and limitations of the technology. It is vital that AI is part of a set of strategic goals. It needs leadership, governance, direction and integration; nothing can be achieved around AI without those, yet many organisations are currently basing key decisions on unrealistic ideas and hype.

AI will undoubtedly usher in significant organisational change in the insurance industry.  This must be considered when tackling the subject of skills gaps and the need for broader understanding of AI use cases from the top down.   Leaders must put in the work to understand the strategic implications of AI, along with its technical capabilities and limitations to manage the change in integrating AI into business strategies effectively. To drive growth and job creation, organisations need to build a model that layers over the top of AI, being powered by it rather than replaced by it.

So how can organisations in the insurance sector begin the daunting task of introducing AI in a measured manner and ensure the best chances of success?

Where to start

The starting point should be in contemplating the alignment of an AI strategy with the business strategy of the company.  The link between the two should be clear and unequivocal.  In addition, aligning with functional, departmental or business unit strategies and plans is also highly desirable, as they should indicate the key objectives of these departments, which should also have already gone through the exercise of alignment with the overall company strategy.  This alignment is essential for the AI team to develop and agree operational plans with the C-suite and other departments.

Establishing value

This process also improves acceptance of the workflow, organisational and process changes that these operational plans bring in practice. Helping the business understand and value all the elements the AI strategy will deliver to teams across the organisation also greatly increases the chances of steering a project to successful completion. 

Ensure universal understanding

All employees need to understand the AI strategy fully, especially where AI is to be employed, why, and the impact it will have on the business.  There is no need for everyone to know all the technical details, but all employees should be able to answer simple questions like “how does my department/team use AI?”, “how do you think AI will impact the way your team/the company works?” and “how does AI impact me in my daily working life – directly and indirectly?”

Workshops and training

Don’t underestimate the complexity of this task.  The ability to answer these questions offers a key metric for the effectiveness of your change management programme and also highlights the need for workshops and training as part of the change management process.  Assuming people have a level of understanding about why AI is being implemented is easily done and can have very negative ramifications, particularly in terms of human resources.

Providing workshops that promote a common understanding of where the company is going with AI and its likely impact allows employees to share concerns and contribute to the strategy. These guided workshops should cover tool selection, AI policy, privacy and security policy and identifying training needs. When done well, this will encourage a feeling of ownership and reduce any anxiety around the change and the introduction of the technology.  Thorough training and clear communication on the AI tools that are available, what they can do and why they were introduced, both before and after the official launch of the strategy will help considerably.  In addition, it should be included as an element of the onboarding process for new employees. 

Identify high impact use cases

The AI strategy should not be overly complicated. The business objectives outlined in the AI strategy should have the potential to deliver significant value but resist the temptation to overload the strategy with objectives that will result in a long-winded and confusing documentation.  

Instead, focus on presenting the AI strategy in a way that will drive collaboration across departments and organisational silos.  As an example, you could use the following business areas and identify possible use cases within each one – Strategy and Vision, Value Realisation, Data Infrastructure and Management, Technology and Tools, Skills and Expertise, Culture and Leadership, Governance and Risk Management, Processes and Operations, Ecosystem, Implementation, Business Impact, and Sustainability.  This is a powerful tool for supporting the creation of AI strategy as it provides a pragmatic and relatively lightweight approach which can be used as a springboard to identifying high impact use cases across business functions.  This forms the basis for the development of your strategy. 

Measure and iterate

Define KPIs to track progress and assess the impact of the implementation of the strategy.  Measure these regularly.  In addition, define what success means upfront – what will it look like?  Create some ‘where are we now’ metrics, then use those as your starting benchmark to measure progress.  Communicating progress regularly across the company with key metrics in an easy to digest dashboard allows your people to see the benefits of your AI strategy rapidly.

Use insights and feedback to refine and adjust strategy over time and encourage constructive feedback.

Build a data-driven culture

Data-driven organisations are going to be more successful at integrating AI than those that lag behind, but it is not too late to start building a data-driven culture.  This is probably the factor that has the most impact in terms of delivering genuine business value and is the most difficult for many organisations to get a grip on.

Data must be accessible, of high quality and be used effectively across the organisation to support AI decision making. The distinction between data engineering and data modelling is key and should be separated into two functions. The models should be made available to everyone and feedback on their effectiveness should be encouraged. This may well be one of the biggest cultural changes for companies – the mindset of an organisation needs to change from hoarding data behind functional gatekeepers (finances, sales, marketing, production engineering and dev, etc.) to making it freely available with the aim of training or enhancing gen AI models.  AI Agents should be encouraged and all employees should have access to them. The success of AI integration lies in the availability of models and agents across the company and the ability and skills of employees to use them as consultants or advisors to become more productive and effective in their roles.

Invest in talent

Investing in AI talent does not mean merely building an internal AI team, but investing in that and combining it with an AI literate workforce who are equipped to use the tools, models and agents provided, and then feed back on their effectiveness.  This is what creates a sense of ownership.

Taking this approach results in a measured, staged creation and implementation of an AI strategy as it makes organisational readiness preparing for AI a major objective and is a key component for any change management process which is necessary to successfully implement AI strategy. 

About the author: James Bent is the VP of Solutions Engineering at Virtuoso QA, a leading innovator in AI-powered test automation. With over 15 years of experience in technology leadership, James Bent spearheads the development and delivery of cutting-edge solutions for end-to-end testing of Enterprise business applications. His thought leadership extends to large-scale talks and workshops at industry events such as Digital Transformation Week, Insurtech Insights, and the National DevOps Conference, together with appearances on podcasts and webinars for Tech and AI industry publications.