Katie Gibbs

Katie Gibbs

Anyone can be upskilled to work in the world of AI as long as they are fast learners and have problem solving skills

Since I started working in the AI space five years ago, the main question that I get asked is around how it will impact jobs. I understand where this worry stems from, and statistics such as “80% of IT jobs will be replaced by AI” are helping to propagate this fear. However, we need to flip the dialogue from a problem statement to an opportunity statement. Rather than focusing on the disruption that it will cause, it’s much better to focus our energy on exploring AI as an opportunity to upskill the workforce and consider what the solutions might look like so we can prepare for this change.

It is often argued that the AI community are discussing these issues within a bubble, that those most impacted by AI will be manual workers – such as drivers – who are disenfranchised from the technology. I’m not sure this holds true any longer. We’ve seen that AI can predict with 79% accuracy the outcomes of human rights trials, so one could make a case that highly qualified workers, such as lawyers, are more likely to be impacted in the immediate future than drivers would. Either way, we need to ensure that we’re engaging with all of those impacted. I strongly believe that, regardless of their current job, anyone can be upskilled to work in the world of AI as long as they are fast learners and have problem solving skills. This is certainly not limited to professions that require a university degree. The question is, how do we reach out to these professionals to provide support? My belief is that it is with governmental policies.

An educational revolution is required to promote life-long learning. In order to achieve this we need to drastically reshape policies to encourage people to learn throughout their career, rather than them viewing it as a one-off activity before they enter the workforce. It isn’t just a matter of finding funding for learning – which can be difficult enough – but it’s also about finding the time to learn in a world we’re everyone is becoming increasingly time poor and where the gig economy promotes longer working hours over personal wellbeing.

It’s worth noting that the University of Helsinki recently opened a free online AI course for to upskill the Finnish workforce. The University aimsto democratise understanding of AI to include a greater proportion of people in discussions around what the future may hold. What they weren’t prepared for was the global uptake of the course, which they highlight in the article. This proves that demand exists for AI education so long as the right training and support is provided. Certainly, in the UK, our government is falling behind in providing support to limit the disruption that AI will have on the workforce.

The reason that I’m focusing on a required change to policies in order to achieve this is because companies are failing to properly invest in providing AI learning to their workforces. While employers should focus on ways to upskill employees on Artificial Intelligence to strengthen their competitive edge by combining brand knowledge with technical capability, we’re instead seeing a large number of firms looking externally to fill AI-specific roles. Not only does this result in disengagement of existing staff who have contributed to growth over the years, but it also means that organisations are paying over the odds for AI specialists – even though they might not necessarily know what they’re looking for in an AI role.

This is why, when looking at the impact of AI on operating models, we take into consideration the employee value proposition. Why should employees engage with AI systems, how does it benefit them and how will their companies invest in training to enable long term value from both sides. There are plenty of transferrable skills that can be used in the design and delivery of AI systems, from upskilling UX designers to design conversational AI that embodies the organisation’s brand and tone of voice for maximum user engagement, to testers who need to learn a new approach to validating AI systems such as image recognition where the results vary widely depending on the data inputted. We’ve seen AI projects fail because they’ve been built in isolation from end users, so when they’ve been thrown over the fence to them, end users don’t engage. Not only does this mean that AI systems are rarely adopted for long enough to achieve RoI, but it is a clear indication of employee’s nervousness when it comes to AI.

The only way to overcome this is through education.

We cannot wait for a new generation of AI-trained professionals to join the workforce – and even if we did, it’s debatable whether that will be the case with the current state of technical education in schools. So governmental policies need to change to support lifelong learning. Only then will employers invest in the future of their workforce by upskilling them to not just cope with the impact of AI, but to embrace it too.