• Written by Muhammed Abdulla NC, CEO

In pharmaceuticals, a sector where 12.4% of US manufacturers reported using AI in the previous two weeks, according to the US Census Bureau's Business Trends and Outlook Survey, the reality of AI adoption is more nuanced than the hype surrounding it. While AI has genuinely changed day-to-day work in some areas, its impact is largely limited to functions whose output is text, rather than those that handle physical work.

Where AI has actually landed in pharmaceuticals so far

AI has mostly been applied to areas such as sales and marketing, where 14.3% of US businesses used AI in the previous six months, according to the US Census Bureau's Business Trends and Outlook Survey. In pharmaceuticals, this translates to tasks such as data analysis, reporting, and document management. For instance, a category manager can use AI to compare two suppliers and generate a report on their performance. However, when it comes to physical work, such as manufacturing and distribution, AI has had limited impact, with only 1.7% of US businesses using AI in production of goods, according to the US Census Bureau's Business Trends and Outlook Survey.

This is because pharmaceuticals involve validated processes under strict documentation rules, making it challenging to implement AI in areas that require physical interaction. As a result, AI has been mostly used to support office functions, rather than operational ones.

Change control overhead

One area where AI has made a significant impact is in change control overhead. When people dealing with change control can ask the data in LIMS, the quality management system, batch MES, and ERP a direct question instead of waiting for a report, it streamlines the process and reduces the overhead. For example, a quality assurance specialist can ask the system about the status of a specific batch, rather than waiting for a report to be generated.

This change has enabled teams to respond faster to changes and deviations, reducing the time spent on manual data retrieval and analysis. By providing direct access to data, AI has reduced the change control overhead, making it easier for teams to manage and implement changes.

Batch record review

Batch record review is another area where AI has made a significant impact. Currently, batch record review involves stitching together data from two systems, which can be time-consuming and prone to errors. With AI, teams can get answers in the moment, reducing the time spent on review and increasing the accuracy of the process.

For instance, a shift supervisor can use AI to review batch records and identify any deviations or issues, rather than relying on manual analysis. This enables teams to respond faster to issues and reduce the risk of errors, improving the overall quality of the batch.

Deviation investigation time

Deviation investigation time is a critical area where AI can make a significant impact. However, it is also an area where AI is often oversold. In reality, deviation investigation time is harder to reduce than it looks, and it requires a thorough understanding of the underlying processes and data.

Before AI can help reduce deviation investigation time, several factors need to be in place, including accurate and complete data, well-defined processes, and a clear understanding of the root causes of deviations. Without these factors in place, AI can add noise to the process, rather than reducing it. As discussed in our research and development efforts, understanding the complexities of deviation investigation is crucial to implementing effective AI solutions.

Why most adopters have not changed anything underneath

According to the US Census Bureau's Business Trends and Outlook Survey, 64.3% of US businesses that have adopted AI have made no change of any kind to adopt it. This means that most adopters have not changed their underlying processes or systems to take full advantage of AI. As a result, they are limited in what they can get out of AI, and their adoption is often superficial.

This is a critical issue, as AI is not a magic bullet that can fix underlying problems. Without changing the underlying processes and systems, AI can only provide incremental improvements, rather than transformative change. As we discussed in our blog post on AI and IT investment in pharma, understanding the need for underlying change is crucial to successful AI adoption.

What a sensible first quarter looks like for a pharmaceuticals business

A sensible first quarter for a pharmaceuticals business looking to adopt AI involves focusing on one question, one system, and one team. This means identifying a specific business problem, such as reducing deviation investigation time, and focusing on solving it using AI.

By starting small and focusing on a specific problem, businesses can build a strong foundation for AI adoption and avoid the pitfalls of superficial adoption. As we recommend to our clients, it's essential to start with a clear understanding of the business problem and the data required to solve it. If you're interested in learning more about how to get started with AI in your pharmaceuticals business, please contact us to discuss your specific needs.

What this does not do, or when it is the wrong choice

While AI has the potential to transform pharmaceuticals, it is not a solution for every problem. In some cases, AI may not be the best choice, such as when the underlying processes are not well-defined or the data is incomplete or inaccurate.

In such cases, it's essential to focus on fixing the underlying issues before adopting AI. As discussed in our analysis of the US Census Bureau's Business Trends and Outlook Survey, understanding the limitations of AI and the importance of underlying change is crucial to successful adoption. By recognizing the limitations of AI and focusing on building a strong foundation, businesses can ensure that their AI adoption efforts are successful and sustainable.