Pfizer is a global pharmaceutical corporation that specializes in the creation, testing, and distribution of medical drugs and equipment. In recent years, the company has introduced AI to aid in the most critical processes, particularly in helping the development of new medicines.
AI in the pharmaceutical industry is more complex, requiring deep learning and pattern identification based on big data (Get Science, 2017). Pfizer is a data-driven organization, and it has data science tools that are transferable as well as can be imported from academic and practice settings, also known as using real-world evidence (RWE) (data from EHR, insurance records) to find expanded use for medicines. AI is also used at Pfizer to digitize clinical trials, adding automation and predictive analytics as well as expanding access to clinical trials by more accurate identification of optimal diversity and location of clinical study sites (Pfizer, n.d.).
Some of the challenges to using the technology are based on the constraints of modern AI, which ultimately relies on competent design and effective algorithms. As the data set and proposed actions grow more complex, there is more room for error. Ultimately, modern AI cannot solve problems or generate new ideas (Greenwald, 2018). From a CRM perspective of human understanding and sales, that means the Pfizer AI continues to be just a supplemental tool for humans, providing recommendations but not being able to resolve complex issues. It can provide recommendations and work within the algorithms to guide b2b relations that Pfizer relies upon to sell its products.
Another challenge and also an issue of ethical and legal nature is the privacy of information. As a drug and medical device manufacturer, Pfizer may have access to private patient data, particularly from trials it conducts and information that is shared by healthcare providers. Healthcare information is highly sensitive and has to be secure at all times, but at the same time, an AI needs large data sets to effectively learn and fulfill its functions which creates a dilemma because it is unclear to what extent will the AI utilize or reveal the private information.
Privacy and data security are hot-button issues regarding AI, and there is no clear consensus on the ownership or security of this data (“Robotics and artificial intelligence,” 2021). Although some regulations exist for commercial applications of data, AI is such a new development and tool, that it is unclear how to ultimately provide oversight.
References
Pfizer. (n.d.). Leading innovation in trials. Web.
Get Science. (2017). A new frontier for AI: Helping scientists develop potential new medicines. Pfizer. Web.
Greenwald, T. (2018). Artificial intelligence (a special report) — What exactly is artificial intelligence, anyway? Everybody’s talking about AI these days. Here’s what all the fuss is about. Wall Street Journal. Web.
Robotics and artificial intelligence. (2021). Gale Global Issues Online Collection. Web.