The New Tech Skills Shortage – AI, Maths, and Statistics
Rapid technological advancements have led to a skills shortage as education and industry struggle to keep up. Over the past decade, there has been a noticeable gap in technical and programming skills among new entrants. This has prompted initiatives and interventions to upskill and cross-train individuals, along with an increase in programming education in schools. As a result, more people now possess programming skills, easing entry into the technical industry. This has been evidenced by the increased number of people with some programming exposure in the training provided by Corecom Technology Academy.
One of the most rapid technological advancements in recent years has been AI and related disciplines. The skills gap for related roles is likely to be much more severe than the general technical and programming skills gap experienced in the last decade.
What is the AI Skills Gap
As the need for organisations to adopt AI increases, so does the demand for related roles, and so does the demand for people with related skills and training. While there is a gap in the general understanding of the use and application of AI in organisations, including for example, developers using Prompt Engineering, the real gap is in implementing and using AI.
The AI skills gap encompasses a range of specialised skills, including machine learning, neural networks, deep learning, and data science. These areas require a strong foundation in mathematics and statistics, which are essential for developing and understanding complex algorithms. Also, there is a growing need for people who can integrate AI technologies into existing systems, optimise performance, and ensure ethical standards are maintained.
The fundamental knowledge and skills that underpin the deployment and use of AI related tech are mathematics and statistics.
Bridging the AI Skills Gap
Bridging the AI skills gap requires concerted efforts from both educational institutions and industry, in the same way efforts were made to bridge the general technical and programming skills gap. But it needs to be faster.
UK Education
Schools must build and expand AI and Data Science curriculum, starting from the foundational topics around maths and statistics. Maths and statistics are already part of the curriculum in UK education, but a focus needs to be put on AI to foster early interest and competence. This needs to be coupled with the increased exposure to programming that students have experienced in the last 10 years. There are relatively simple interventions and initiatives that could be implemented in schools and universities that bring together the existing curriculum around mathematics and programming, giving students the opportunity to experience simple AI solutions, like chatbots or image matching, that will spark interest and further learning.
Tech Industry
Industry can support continuous learning and professional development by offering training programs and certifications that keep pace with technological advancements. This needs to start with the fundamentals – mathematics and statistics. Many people already working in the industry, in general tech roles or beyond, will have quickly forgotten the key concepts in maths and statistics required to begin learning AI. Indeed, some may not have covered these concepts at school or university at all. For Machine Learning specifically, topics like Algebra, Calculus, and general probability and statistics have probably rarely been used in adult working life. For existing people in industry, an initial training focus on mathematics and statistics needs to be employed to give people the opportunity to begin a learning pathway in AI.
Career Switchers or Junior Talent
Beyond a solid foundation in mathematics and statistics, people wanting to enter the fields of AI and machine learning (ML) need to gain expertise in several key areas. These include understanding:
Machine learning algorithms
Neural networks
Deep learning frameworks
Practical experience with programming languages such as Python or R, along with proficiency in using tools like TensorFlow, PyTorch, and Keras, is essential. Training should also encompass data preprocessing, model evaluation, and deployment techniques. Also, knowledge of cloud computing platforms like AWS, Google Cloud, or Azure can be highly beneficial for implementing scalable AI solutions. This amounts to several weeks of training, which can be disruptive to people’s existing project work, but can be managed through staged training programmes over several months, or by using specialist providers that provide people already trained in the specific disciplines required by a given organisation.
By aligning various educational Learning Pathways with industry needs, we can create a skilled workforce ready to meet the demands of the evolving AI landscape, but this needs to start with a focus on mathematics and statistics as the underlying skill that underpins everything.
Author: Thomas Crabtree