Xavier Fernandes, Analytics Director at Metapraxis
A study by Oxford University academics into “The Future of Employment” in 2013 prompted apocalyptic headlines which stated that in the future 40% of jobs will be automated thanks to advancing technology.
The researchers subsequently claimed that the truth was in fact a little more prosaic; rather than facing complete automation, the research found that 40% of jobs faced some aspect of automation in their activity. So with new ‘AI processes a likely reality for almost half us, what does that mean for our current roles and should we be worried?
The fourth revolution?
The first industrial revolution saw machines replacing muscle, both human and animal. The second and third saw electrical power, mass production and computerisation revolutionise the job market. Now, with daily headlines of AI as an employment superpower, there is some concern that AI is bringing a fourth revolution, and with it, unknown circumstances.
This ‘fourth industrial revolution’ is defined by replacing brain power with machines. Our thinking capacity is what inherently sets us apart from other species, so it’s not surprising that any encroachment on it triggers some existential angst.
Evolve to reap the rewards
While many businesses still don’t fully understand the capabilities of AI, those who fear its development are, instead of embracing it, missing all the benefits that it can bring to the workplace. Businesses that utilise AI appropriately are seeing vast improvements across their entire value chain; better customer experience, reduced costs, and more insightful analysis to support management decisions.
AI is particularly useful for supporting tasks with repetitive activity, for example, performing financial checks and assessing large sets of data within financial services firms. AI performs particularly well within this context, spotting outliers before a human expert would notice them, allowing impending problems to be flagged and avoiding costly mistakes.
There is also an increasing focus on maximising customer lifetime value through the use of AI. Being able to predict existing customers’ needs as well as track trends in their financial circumstances is supercharging the old cross-selling approach with testable, predictable outcomes.
With potential benefits like these on offer, management teams of innovative financial services are increasingly relying on AI to help them with some of the heavy-lifting of analysis. Using advanced data capabilities and learned behaviours, AI analyses market trends to provide predictions of future performance. This insight is invaluable and allows management teams to change direction and correct any problems accordingly. This offers a huge advantage over those that have not adopted such tools.
Supporting the workplace
Algorithms and AI are typically ‘smart’ at doing one, tightly-constrained task, but they can be less helpful with many of the activities that humans find straightforward. In most white-collar jobs, automation tends to replace certain tasks in the job, rather than the role in its entirety, as the need for human intelligence is still highly necessary. In particular, we still need human input to first challenge, and then synthesise, this information before taking action. Employees should therefore work with the business to proactively identify what areas of their role could be automated, so that they can focus on the areas that add real value to the business’ commercial goals.
Challenging AI is certainly still important. We know that algorithms can be much better than humans on certain, bounded tasks. However, many algorithms rely on existing data sets to build their understanding. As a result, when a business unit has ‘symptoms’ that fall outside of that body of knowledge, the algorithm may suggest the wrong course of action with costly results.
Indeed, even with plenty of data, algorithms will reflect any biases the data set contains. We’re seeing this with some legal sentencing algorithms where there is evidence that they are treating disadvantaged people more harshly. Getting the answers to why and how far we should trust our algorithms should therefore become an everyday part of any job affected by AI.
Rather than depending entirely on AI for all decisions, workers should be taking all these new, AI-generated insights and using them to complement the human decision-making process. No manager of a complex business ever has enough time to sieve through all the analysis available, but with AI driven algorithms able to flag up any issues and indicate where action needs to be taken, we may find that we have some AI ’colleagues’ who will cover our backs and suggest innovative options. Yes, there will be times when the algorithms get it wrong, but as long as we’re watching out for those, the future is bright.
