The most significant impact of AI won’t be on the number of jobs, but on job content.

Today Tomorrow
A risk and operations professional manually updates and checks various types of compliance reporting and controls for a business or functional area. AI platforms, using machine learning and predictive analytics, simplify and drive efficiency in data gathering, raise the quality of controls, augment the risk & ops professional, and free up time for her to focus on analysis, an end-to-end view of the organisation’s risk profile, and early identification and rectification of issues.
A financial advisor spends a significant amount of time onboarding a potential customer, taking personal details and interrogating his financial situation. Then she goes away to do the research and hopefully secure the relationship. The prospective customer goes online and uses AI to onboard himself. He completes the other administrative requirements and provides relevant research which he has sourced. This allows the financial advisor to focus on the true value of her offering: building the relationship and providing quality advice.

Machines will amplify the capabilities, effort, and impact of humans by augmenting their intelligence. But humans will also improve the performance of intelligent technologies, acting as trainers (assisting computers to learn), explainers (interpreting and explaining the decisions of AI to customers) and sustainers (ensuring intelligent systems stay true to their original goals without crossing ethical lines).

FS needs a systematic approach to unlock AI value

While just over half of all banking and insurance executives acknowledge that getting human-machine collaboration right is critical to achieving their goals, few have adopted a systematic approach to unlocking that value.

Training is critical, but a lack of clarity on the skills to prioritise is a major inhibitor. Moving the spotlight from jobs to the nature of the work itself is an important first principle to adopt.

When reconfiguring work, FS firms need to take three steps:

Assess tasks, not jobs—Identify the new kinds of tasks that must be performed and allocate them to people or machines. Identify those tasks that fit the ‘missing middle’ (see Figure 1)—those tasks best done by machines and humans in collaboration. The process of allocation of tasks will be ongoing and initial errors and unintended consequences are likely. Constant observation and corrections to initial allocations will thus be important.

Figure 1. The ‘missing middle’—the tasks best done through man-machine collaboration

Create new roles—AI enables people to take on higher-value work so FS organisations will want to set up new roles within a broader context. For example, operational jobs will become more insight-driven and strategic, while mono-skilled roles will become multi-skilled. Jobs will also become more specialised and new jobs will emerge—e.g., human trainers to teach AI chatbots about tone and empathy.

Map skills to new roles—with a list of required tasks, skills and newly defined roles, the organisation can map what is required against the skills present in its workforce. The gaps can be addressed through training or sourcing. In creating new roles, FS organisations should focus more on how the work can best be done than on the roles and responsibilities that make up the work. The reconfiguration of roles should be a flexible process that looks beyond full-time equivalents and considers a range of possibilities, including build, buy, borrow and bot. This will enable skills to be redeployed as and when needed.

Join me next week as I look at how FS organisations can pivot the workforce to areas that create new forms of value.

Meantime, for more insight on the impact of AI, see Accenture’s latest research and thought leadership:

Submit a Comment

Your email address will not be published.