Skip to main content

Training and tech: how to solve the AI skills gap in a post-pandemic age

machine learning
(Image credit: Image Credit: Geralt / Pixabay)

From pharmaceuticals to retail, AI is changing the face of almost every major industry sector right now – in a tectonic shift that’s snowballed as a direct result of the pandemic. Over half of US companies are ramping up their AI investments because of Covid-19, in a trend echoed globally. This includes the UK, where the impact of AI technologies is expected to drive a GDP gain of at least 5 percent by the year 2030. 

However, the success of AI, and the larger vision of Industry 4.0 that it plays into, depends on a broad coalition of data scientists – along with the right environment in which their innovations can thrive. With the World Economic Forum predicting the creation of 97 million new jobs due to AI in the next four years, this booming commercial opportunity is fast becoming something of a juggernaut. However, to take advantage of this promising new technology, businesses need AI-trained IT, science and engineering teams. 

The AI talent bottleneck is a time-honored problem that has only intensified due to Covid-19, and the subsequent increase in data-optimized development processes. If the UK is to achieve its ambition of becoming a global AI superpower in the next decade, it must first address the major roadblock of AI training and talent retention.

According to a wide-ranging Microsoft report released last year, the UK is in the grips of a major skills crisis: just 17 percent of employees are being re-skilled for AI, compared to 38 percent globally. Meanwhile, 52 percent of employees are using AI to work faster and smarter, compared to 69 percent of employees globally. 

In this fast-moving global economy, the UK’s business leaders must take dramatic action not to lose their foothold and maintain a competitive edge.

Fixing the talent pipeline

The race for accelerated digital maturity is a challenge that, in part, is ideally addressed from the bottom up, but that comes with its own challenges. For instance, despite the digital literacy of generation Z, the number of young people taking IT subjects at GCSE has fallen by 40 percent since 2015. This dip, mirrored at A-Level and further education levels, has many business leaders concerned that young people are leaving full-time education without sufficient advanced digital skills. 

To inspire the next generation to pursue courses and careers in IT and STEM requires a multifaceted approach including new and engaging curriculum development, greater access to edtech and reskilling teachers – all of which are central to shaping a future generation of AI talent, including more women, to the sector. Initiatives such as STEM Learning, which works with the government to bring STEM role models into schools, are making great progress in driving the early STEM uptake that will be crucial to creating long-term change.

While important, however, this is only the first in a series of intrinsic barriers that are stymying AI talent development. Take the tech world’s well-documented diversity crisis. This “disastrous” situation not only replicates gender and racial biases at a product level in AI, it’s also a self-perpetuating problem. The fewer women or people of color there are at leadership level in a field that is fast dominating the systems we live by, the less candidates from marginalized groups will be attracted to, and stay, in the sector.

As well as looking at the pipeline that leads potential employees from school to the industry, therefore, businesses also need to consider obstacles such as harassment or promotion bias that are further complicating AI talent retention.

Using AI to maximize what you have

With AI becoming more ubiquitous by the day, it’s worth thinking internally about how adopting this technology can help plug the industry’s endemic talent gap. Just as machine learning is key to agile working processes, a business model with AI at its core empowers a workforce to pivot quickly to changes and demands. 

Rather than seeing AI as a niche qualification, businesses should consider the broader skills – problem-solving and an analytical mindset, for example – that could be nurtured by rechanneling the talent they already have. When managed correctly, this strategy could also help with the democratization of AI that is vital to innovation, helping teams stay one step ahead of cutting-edge tech.

Creating an internal culture of development is also a good mechanism for accommodating the crossover between AI and engineering that nearly all sectors will encounter. According to Vincent Higgins, global director at tech futurist company Honeywell, “The most common mistake people make is that they hire data scientists without bringing the subject matter experts along. Successful application of AI is a marriage of data and expertise right down to the granular level.”

With Britain facing a “retirement cliff” in engineering, this is another sector that, like AI, faces an imminent talent shortage. But AI can also bridge the experience gap in engineering, with the creation of a predictive AI model that is built for engineering purposes. 

With the right R&D processes in play, businesses can bring together their existing data lakes (explicit data) with the knowledge and complex physical behavior of seasoned engineers (implicit data). This, in turn, will leverage all their capabilities across data science and engineering, supporting a new generation of employees to optimize designs of the future in a talent-efficient way. 

Innovation: the cause and the cure

The hyper-digitalization we’ve seen in the past 18 months has aggravated the AI talent gap, and the issue is twofold. Firstly, there is a longer-term game plan that needs to be enacted, to address issues such as STEM uptake in schools, and the lack of progression opportunities at the beating heart of tech. 

These problems will take time to solve, which is why it’s also worth thinking about a series of short-term opportunities whereby the problem becomes the fix. First, with the right historical data embedded into their engineering workflows, businesses can build powerful machine learning models that cut down on iterations, and accelerate R&D processes. 

By building AI solutions collaboratively (opens in new tab), businesses can build a walkway between engineering experience and computer programming, in a foundation that makes the most of their in-house capabilities. Using a strong set of AI tools, engineers can run simulations in batches (for new design concepts) or even eliminate prototyping altogether (for evolutionary designs) for a more responsive system that retains inter-generational knowledge. 

Rather than relying on a limited pool of outside talent, this smart approach harnesses existing skills. The result is a dynamic, AI-literate team that walks the perfect balance between computer programming and human experience; and a business model that will roll with the punches ahead.

Dr. Richard Ahlfeld, CEO and founder, Monolith AI (opens in new tab)

Dr. Richard Ahlfeld

Dr. Richard Ahlfeld, CEO and founder, Monolith AI.