Introduction
“…the most profound technology humanity is working on. More profound than fire or electricity or anything we’ve done in the past” – Sundar Pichai, CEO of Alphabet and Google (Zubrow, 2023, Pichai, 2023).
Artificial intelligence (AI) appears to be everywhere and is increasingly becoming a staple of modern life, which begs the question, are the life sciences ready for more AI?
AI is a powerful technology that has the potential to revolutionise business operations and, understandably, has become increasingly used within the life sciences industry, worth a staggering USD 2.83 trillion (Deloitte China, 2023). The earliest uses of AI in the life sciences date back to the 1960s when pioneering Dendral experiments were conducted with a computer system to predict the structure of molecules. The experiments increased interest in using computers within the life sciences and influenced the development of subsequent systems such as MYCIN (created in 1972). MYCIN was used to identify the bacteria that caused infections in patients and is acknowledged as one of the earliest applications of AI in medicine (Patel et al., 2008; Shortliffe, 1977). Fast forward to 2023, the market value of AI within the life sciences was valued at over USD 2 billion (Mordor Intelligence, 2024), with utilisation across a highly diversified set of roles, including bioinformatics, biomedical engineering, data analysis, drug discovery and development, quantitive biology and systems biology. Moreover, the utility of AI was displayed throughout the COVID-19 pandemic, as it was used to model the risk of outbreaks, for epidemiological analysis (Arora et al., 2020), and even to design the vaccines used (Sharma et al., 2022).
The impact of AI is undeniable, and this extends to the job market, with approximately 97 (Ascott, 2021) and 85 million (World Economic Forum, 2020) jobs estimated to be created and lost respectively by 2025. While the granular details on the scale and type of impact AI will have on the life sciences job market are still yet to be revealed, current trends show that job opportunities may grow, with the compound annual growth rate (CAGR) of AI in the life sciences expected to increase at 25.2% between the forecast period of 2024-2029 with a market value of USD 8.88 billion by the year 2029 (Mordor Intelligence, 2024).
With the use of AI increasing within the life sciences, educational and commercial organisations must be involved in the training and recruitment of digital talent to prevent the development of a skills gap, which could increase sector inefficiency. This article will present a view on whether the current and prospective workforces for the life science sector are ready for AI, exploring:
- How educational and commercial institutions nurture digital talent
- The existing skills gap and how they can be addressed
Education is vital – the current landscape
Education is vital for providing digital talent with the skills to effectively use AI in the life sciences. What, then, is the current outlook?
For school-aged children (typically aged between 5 and 18), there appears to be minimal evidence of schools teaching AI as a specific subject. However, there are noteworthy cases where organisations have developed curriculums for schools. One such case is the Erasmus project, which started in 2019 and is entitled “AI + : Developing an Artificial Intelligence Curriculum Adapted to European High School” (Bellas et al., 2023). A curriculum was developed and piloted across six schools from five European countries: Finland, Italy, Lithuania, Slovenia and Spain. The purpose of this curriculum is to help school-age children better understand how AI works and the implications of its use in the real world. Another case is in India, where the Central Board of Secondary Education (CBSE) introduced an AI curriculum for grade 9 students (14-15 years of age) as an optional subject (CBSE, 2019; Chiu et al., 2021).
Although much still needs to be done to integrate AI education in schools, governmental bodies know its growing importance, as shown by the ethical guidelines for AI in education published by the European Commission (Directorate General for Education, European Commission, 2022) and the insights and recommendations for AI in teaching and learning published by the US Department of Education (US Department of Education, 2023).
Conversely, there seems to be more AI education in universities, with a reported 80% of surveyed academics from several European countries teaching AI in courses related to smart networks (Dec et al., 2022). Specifically, current data shows that 109 institutions in the UK offer AI courses (IDP Connect, 2024a), while the US has 69 institutions offering AI courses and 43 providing postgraduate studies (IDP Connect, 2024b,c). Moreover, examples of tailored AI education for the life sciences include the Artificial Intelligence in the Biosciences MSc, offered at Queen Mary University, London; MSc AI for Biomedicine and Healthcare, provided at the University College London, and Artificial Intelligence for Biomedicine and Healthcare offered at the University of Torino.
The growing provisions of AI education in universities (and at an advanced level) could broaden the digital skills gap between the next and current generation of workers. Employment training is an area that may be lacking, given that only 13% of workers globally were offered AI training in the past 12 months, despite over half of them thinking that AI skills are essential for their jobs (Randstad, 2023).
Overall, there appear to be significant gaps between AI education at the school-age level and university and between university and those currently employed.
What challenges could impact the delivery of AI education for the life sciences?
Challenges in providing AI education for the life sciences are multifaceted. One core dilemma is societal readiness, which, if not addressed, may hinder the ability to scale AI use effectively, particularly in pre-university education. This view is implied by:
- Research which shows that teachers may benefit from increased support for AI education, including expanded or refined training (van Brummelen and Lin, 2020)
- The fact that AI in education is relatively new means there is still much to learn and standardise where best teaching practices are concerned (van Brummelen and Lin, 2020; Kim et al., 2021)
- Socioeconomic concerns that AI may increase the inequality between well-funded and underfunded schools by providing those from a wealthier background a better learning experience (Klein, 2020)
Additionally, scepticism around AI in the workplace could prevent the use of the technology and employment-based training. Key concerns include:
- Data quality and the inability to determine how AI models reach their conclusions (Paul et al., 2021; Lamberti et al., 2019)
- Concerns of job loss
- Cost of resources – in the UK, 13% of small and medium-sized companies mentioned that costs were a barrier to AI implementation (Capital Economics and DCMS, 2022)
Finally, the skill gaps for AI will impact the ability to nurture digital talent, and the consequences could be far-reaching due to a delay in adopting technologies capable of improving the sector’s efficiency. McKinsey’s estimate that AI could generate between USD 60-110 billion in value for pharmaceutical and medical product businesses suggests AI’s potential economic value for the sector and why quick adoption would be ideal (McKinsey & Company, 2024).
Bridging the gap
In 2020, McKinsey reported that 80% of pharmaceutical companies mentioned a ‘skills mismatch’ attributed partly to ‘digitalisation’ and ‘advanced analytics’ (McKinsey & Company, 2020). In leading markets, i.e., 14 OECD countries across North America, Europe and the Pacific, the OECD-Lightcast report showed the demand for AI skills increased by 33% (OECD, 2023).
According to statistics from the UK, an excess of 133,000 jobs in the life sciences will need to be filled by 2030. Computational roles such as physiological modelling, computational chemistry, pharmacokinetic and pharmacodynamic modelling were cited (SThree, 2022) as areas where there needs to be more recruitment. Whilst there appears to be limited data on the specific amount invested to address the skills gap, there’s evidence that could suggest the past and current efforts of commercial and governmental institutions are working towards addressing this dilemma:
- Deloitte reported that more than 40% of life science companies invested between USD 20-50 million in AI initiatives (Deloitte Insights, 2020)
- There is ongoing collaboration between government and industry to increase opportunities for skills development (Science Industry Partnerships, 2020)
- Direct investment: a keynote example is the UK, where since 2014, over £250 million has gone towards AI development in the life sciences (Financier WorldWide, 2023)
Are the life sciences ready for more AI?
AI technology is a powerful technology that is still in its infancy, impacting the degree to which educational programmes can be provided, particularly at the pre-university level. The structures to nurture digital talent for the life sciences are far more robust in universities and beyond. However, challenges, mainly related to social readiness and the current quality of AI technology, could impact the scale of training provisions. Although limited data is available regarding AI’s use within the life sciences, these insights shed light on the long developmental road ahead.
With the advent of the 4th industrial revolution, AI will become increasingly used within the life sciences. To ensure that the industry can capitalise on the many benefits of AI use in the sector, educational, governmental and commercial organisations must work together to address the skills gap for digital talent.
