AI (Artificial Intelligence) has increasingly become integral to our lives. Several job sectors and companies have already incorporated AI into their daily operations, and Life Sciences is no exception. However, along with AI comes a myriad of ethical considerations that must be addressed to ensure everything is above board. Unsure about how to approach the use of AI? Read on below.
In 2023, McKinsey published a report on the future of generative AI, predicting that generative AI alone has the potential to increase revenue in the pharmaceuticals and medical products industry by 2.6-4.5% of the industry revenue as a whole. This positioned the industry as the third-highest potential for AI-driven growth, following high tech and banking (See figure from the McKinsey report):
(McKinsey, 2023: p. 7)
McKinsey highlights that AI can have significant impact and utility in the areas of research, drug discovery, software engineering, and marketing and sales in the Life Sciences industry if integrated properly, see below:
(McKinsey, 2023: p. 8)
However, there are several ethical considerations and concerns that immediately arise from the AI use cases outlined across the value chain. Consulting firm WittKieffer identifies five primary AI ethics concerns in their article, including:
Disinformation
Generative AI can create content that appears coherent and accurate by analyzing patterns in large datasets rather than understanding the actual meaning of words. However, this content can often be misleading, contradictory, or entirely incorrect, as it may not necessarily reflect reality. Therefore, ethical considerations regarding the production of disinformation through the use of AI should be taken into account.
Privacy
When discussing AI, it's hard to avoid the conversation about privacy. Many AI models continuously train themselves using the queries and data presented to them, which can increase the risk of privacy breaches. Be cautious about providing overly confidential data, as we have yet to understand the potential risks fully. While the use and reuse of patient data offer significant opportunities for research, it also raises challenges regarding privacy, data ownership, and protection.
Bias
Bias is described as nearly unavoidable within the field of AI ethics because AI programs often learn bias from the data they consume rather than from the programmer. This can be particularly problematic in bio-scientific applications, where a non-representative group of participants in a clinical trial may be used to train models or evaluate the viability of treatments, leading to results specific to a particular group.
Environmental Concerns
Lastly, it is pointed out that AI can pose environmental challenges, including significant carbon footprints from the computer and information storage sector, as well as carbon dioxide production from AI model training. WittKieffer points to a research from Massachusetts that shows that AI model training can produce around 626,000 pounds of carbon dioxide – equivalent to about 300 round-trip flights between New York and San Francisco.
Conclusion
The integration of AI into the Life Sciences industry holds vast potential for revolutionizing research, drug discovery, and various operational aspects. However, this technology also brings ethical considerations that demand careful attention and action. From the potential spread of disinformation to privacy breaches, biases, and environmental impacts, the ethical implications of AI are complex and far-reaching.
While McKinsey's report highlights the immense growth opportunities that AI presents, it is essential to approach its integration with a critical eye towards ethical implications. WittKieffer's identification of key AI ethics concerns underscores the necessity for proactive measures to mitigate risks and uphold ethical standards in AI-driven applications within the Life Sciences sector.
Click here to read our article on: "Leading organizations”
Click here to send us your CV.
Sources:
The Ethics of AI in Life Sciences: How leaders must navigate trust and transparency. (2024, March 6). WittKieffer. https://wittkieffer.com/insights/the-ethics-of-ai-in-life-sciences-how-leaders-must-navigate-trust-and-transparency
What’s the future of generative AI? An early view in 15 charts. (2023). In McKinsey & Company.