Will AI doctors disrupt drug commercialization?

Key points

Artificial intelligence and large language models are set to revolutionize how pharmaceutical companies handle data, information, and relationships, potentially transforming traditional disease diagnosis and treatment.
Pharmaceutical companies spend USD 30 to 50 billion annually on drug marketing, with the largest portion aimed at healthcare professionals, followed by direct-to-consumer and digital marketing.
Advanced AI-driven medical chatbots have shown remarkable diagnostic accuracy, with studies indicating a 90% accuracy rate for AI chatbots alone, compared to 76% for AI-assisted physicians.
The integration of AI into drug commercialization could disrupt traditional roles, with AI-powered e-detailing posing a threat to sales representatives and potentially reshaping physician-patient relationships.

Drug commercialization or pharmaceutical marketing industry is vast, yet it often operates behind the scenes. Historically, pharmaceutical companies directed their marketing efforts toward two primary audiences: healthcare professionals and patients. These efforts can take various forms, including face-to-face interactions, traditional media, social media, and digital platforms. However, the rapid rise of artificial intelligence (AI) and large language models (LLMs) is poised to lead to new ways of how data, information, and relationships are managed, ultimately transforming the traditional model of disease diagnosis and treatment.

The current state of drug commercialization

Globally, pharmaceutical companies spend between USD 30 to 50 billion annually on drug marketing,¹ primarily for newly launched drugs. Marketing expenses tend to gradually diminish as patents expire. These expenditures mainly fall into three major categories:

Marketing to healthcare professionals (USD 20–25 billion)
The largest share of the overall marketing budget is dedicated to promoting drugs to healthcare professionals. They usually involve sales representatives who distribute brochures, clinical study reprints, and drug samples during face-to-face meetings with healthcare professionals. However, the effectiveness of these interactions is often constrained by the limited time physicians can spare, with sales rep visits typically lasting only a few minutes.
Direct-to-consumer marketing (USD 6–8 billion)
While over-the-counter (OTC) drug advertising is permitted in most countries, prescription drug marketing is heavily regulated and allowed on a large scale only in the US. This sets the US apart from the rest of the world in terms of new product launch and adoption as well as medical devices innovation. The practice is not without controversy. Critics argue that such marketing drives up healthcare costs and misleads patients, while proponents contend it raises disease awareness and reduces the stigma associated with some chronic diseases. Success stories of direct-to-consumer marketing include Pfizer’s blockbuster erectile dysfunction drug Viagra, even after its patents expired in 2020, Merck’s immunotherapy drug Keytruda for lung cancer patients, and AbbVie’s rheumatoid and psoriatic arthritis drug Humira.
Digital marketing (USD 3–5 billion)
The COVID-19 pandemic accelerated the adoption of digital marketing in the industry. During this period, people explored telemedicine, online webinars, and social media, experiencing some success. However, as life returned to normal, these platforms quickly failed to deliver on the grandiose ambitions set during the peak of the lockdown. This time, however, the emergence of LLM appears to be different. These AI doctors are beginning to sound like real doctors who can understand human language, not just computer codes.

Are AI doctors better at diagnosing patients?

The concept of AI-driven medical chatbots is not new, but the emergence of advanced LLMs like ChatGPT has elevated diagnostics accuracy and conversational capabilities to unprecedented levels. Some believe the chat interface has been the game changer.² While these tools are still used on a limited basis by physicians so far, there is increasing evidence that their potential to enhance healthcare productivity is immense.

A research paper published in JAMA in November 2024³ underscores this potential (see Figure 1 for the study flow). Initially designed to compare the diagnostic accuracy of physicians using AI versus those using conventional resources such as search engines, the study yielded a surprising finding: while the accuracy of AI-assisted physicians (76%) was only marginally higher than those assisted by conventional resources (74%), the AI chatbot alone achieved a remarkable 90% accuracy rate (see Figure 2).

Figure 1: Study flow diagram

Source: Goh E et al. JAMA Network Open (n.d.). eFigure 2: Study Flow Diagram. Page 16. Retrived from on 25.03.2025

A schematic diagram showing the organization of the study, with 25 physicians assisted by GPT and 25 using conventional resources to complete six diagnostic cases within 1 hour.

Figure 2: Median diagnostic reasoning score

Three pie charts depicting accuracy of AI-assisted physicians vs of those assisted by conventional resources

Source: The Nex York Times (2024). A.I. Chatbots Defeated Doctors at Diagnosing Illness

Three pie charts depicting 76% accuracy of AI-assisted physicians vs 74% of those assisted by conventional resources compared to 90% of LLM alone.

Interestingly, the study pointed out that physicians are somewhat reluctant to let AI guide their thinking process, even though LLMs have demonstrated remarkable human-like reasoning capabilities. Arguably, AI should perform even better in cases involving vast and complex data that surpass processing capabilities of the human brain.

Shifting patient and physician attitudes

Previous studies have shown that patients were reluctant to trust AI diagnostics.⁴ However, this could change quickly if more studies demonstrate that AI can deliver higher diagnostic accuracy. It remains unclear how this shift might affect the traditional physician-patient relationship.

For sales representatives, AI-powered e-detailing could pose a real threat. Imagine an AI sales rep who knows every single detail about a drug (or any drug) and can answer a doctor’s inquiry at any time of the day. In the past, the pushback has always been the personal touch and the long-term trustworthy relationship between the doctor and the sales rep, but AI would allow doctors to be better equipped facing the reps.

AI changing the dynamics of drug commercialization

The integration of LLM-based AI chatbots into drug commercialization is likely to have a more profound impact than previous digital tools. Major stakeholders – physicians, patients, and drug companies – will continue to adopt AI in their daily workflows and achieve significant productivity gains. The convergence of AI and healthcare is poised to create a paradigm shift that redefies the roles of stakeholders in the healthcare ecosystem and consequently reallocate revenue and profit share in global drug commercialization. We strive to identify companies that are at the forefront of capturing the enormous pharma marketing budget shift and remain excited to witness the birth of novel business models.

^1.Kantar (n.d.). Great pharmaceutical advertising: It ain’t easy. Retrieved from on 28.03.2025; JAMA Network (2019). Medical Marketing in the United States, 1997-2016. Retrieved from on 28.03.2025
^2. The Nex York Times (2024). A.I. Chatbots Defeated Doctors at Diagnosing Illness. Retrieved from on 26.03.2025
^3. Goh E, Gallo R, Hom J, et al. (2024). Large Language Model Influence on Diagnostic Reasoning: A Randomized Clinical Trial. JAMA Netw Open. 2024;7(10):e2440969. doi:10.1001/jamanetworkopen.2024.40969. Retrived from on 24.03.2025
^4. National Library of Medicine (n.d.). Diverse patients’ attitudes towards Artificial Intelligence (AI) in diagnosis. Retrieved from on 26.03.2025

S-04/25 M-001107

About the author

Fang Liu
CFA, Portfolio manager, Thematic Equities
Fang Liu is a senior portfolio manager for the Digital Health Equity strategy on the Thematic Equity team at ��ţ�� Asset Management. Before joining the team in February 2020, Fang worked for 3 years in the equity investment team at Calibrium AG, managing a few global all-sector concentrated high-conviction strategies. Prior to that, she worked for Lombard Odier as an equity analyst in the thematic team since 2015. Fang spent 4 years as an academic researcher at IMD business school, where she acquired comprehensive research skills and broad industries and sectors knowledge. Fang holds a master’s degree in Management from the University of Lausanne (HEC) and is a CFA Charterholder, a member of the CFA Institute and the CFA Society of Zurich.