AI in Drug Discovery: Revolutionizing Pharmaceutical Research

Of all the ways artificial intelligence (AI) transforms our lives, perhaps the most exciting is its potential to revolutionize medicine by accelerating drug discovery. Historically, the pharmaceutical industry has focused more on treatment than cures. Can you recall the last time a disease was truly cured or eradicated? In eye care, for instance, while treatments for conditions like macular degeneration and glaucoma have advanced, true cures have yet to be discovered.

However, the integration of AI into pharmaceutical research is beginning to shift the landscape. With it, is a renewed hope for providers and patients alike. By streamlining clinical trial design and accelerating the identification of novel therapies, AI is opening the door to faster, more effective and potentially curative treatments.¹

The Challenge of Traditional Drug Development

Bringing a new drug to market is a notoriously lengthy and expensive process. From initial discovery to regulatory approval, the process can span 10 to 15 years and cost upwards of $2.6 billion.¹ Of the thousands of compounds screened, only about 5–10% make it through clinical trials to market.²

For smaller biotech pharmaceutical companies, a failed trial can be financially devastating, often leading to bankruptcy. A recent example is Oxurion, a company focused on retinal diseases, which filed for bankruptcy after its Phase 2b trial for diabetic macular edema failed to meet its primary endpoint.³ In contrast, larger firms such as Pfizer, Eli Lilly and Roche are typically better positioned to absorb such losses.

Even when drugs do reach the market, they often debut with a hefty price tag, primarily to recoup the high costs of research and development. Despite the resources and expertise available to biotech and big pharma companies, traditional clinical trials remain fraught with inefficiencies. These include under-enrollment, high dropout rates, manual data collection and rigid protocols that struggle to accommodate the complexities of real-world patient populations.¹ Such limitations not only delay therapeutic innovation but also contribute to the high failure rates and escalating costs that plague the industry.⁴

AI: A Transformative Force in Drug Development

AI offers a powerful solution to many of these challenges. By rapidly analyzing vast datasets, including molecular structures, clinical trial outcomes and patient health records, AI can identify promising drug candidates with greater speed and precision than traditional methods.⁵ Its ability to learn from historical data and simulate biological interactions enables researchers to make more informed decisions at every stage of drug development.

Target Identification and Preclinical Research

One of AI’s most impactful applications is target identification. AI algorithms can sift through multiomics data, literature and molecular databases to pinpoint biological targets with high therapeutic potential. This reduces the reliance on trial-and-error experimentation and increases the likelihood of success in early-stage research.⁵

Open-source tools like the Open Drug Discovery Toolkit (ODDT) and SAveRUNNER exemplify how AI is democratizing access to advanced drug discovery capabilities. SAveRUNNER, developed by Italy’s National Research Council, uses network-based analysis to identify new indications for existing drugs, accelerating the repurposing process.⁶

Revolutionizing Clinical Trial Design

AI is also reshaping how clinical trials are designed and executed. For rare diseases such as retinitis pigmentosa, recruiting enough participants is a major hurdle. AI can analyze epidemiological data to identify geographic regions with higher concentrations of eligible patients, optimize inclusion/exclusion criteria and determine ideal sample sizes.⁷

During trials, AI can monitor patient adherence, detect outliers and suggest mid-study adjustments where applicable/permissible to improve outcomes. It automates statistical analysis and generates visualizations that support regulatory submissions, reducing the burden on researchers and improving data integrity.⁷

Platforms like ACUITY, developed by Cancer Research UK’s digital Experimental Cancer Medicine Team (ECMT), are already being used in multiple clinical trials. ACUITY integrates real-time data from various sources, enabling researchers to make faster, more informed decisions and reduce reliance on proprietary software.⁸

Enhancing Patient Engagement and Personalization

AI’s potential extends beyond trial logistics to improve patient experience and personalization. Tools like PROACT, part of the ACUITY system, allow patients to report outcomes via video, audio or text. These insights help researchers understand how treatments affect patients’ daily lives, leading to more patient-centric trial designs and therapeutic strategies.⁹

AI also facilitates the development of personalized medicine by analyzing genetic, phenotypic and behavioral data. This enables the creation of tailored treatments that are more effective and better tolerated, especially for complex or chronic conditions.

Industry Adoption and Strategic Collaborations

Major pharmaceutical companies are increasingly investing in AI-driven research. AstraZeneca partnered with BenevolentAI to identify novel targets for idiopathic pulmonary fibrosis and chronic kidney disease, resulting in multiple promising candidates.¹⁰ Takeda,Bristol-Myers Squibb (BMS) and Medidata AI have also launched collaborations to integrate machine learning into their drug development pipelines.¹¹

Repurposing and Accelerating Drug Discovery

AI is proving especially valuable in drug repurposing, the process of finding new uses for existing medications. This approach can significantly reduce development time and costs. For example, PandaOmics and Iktos use AI for generative chemistry and predictive modeling, helping researchers identify new therapeutic applications for known compounds.¹²

Biosimulation platforms like VeriSIM Life’s BIOiSIM use AI to simulate drug interactions within virtual human systems. This allows researchers to test efficacy and safety before entering clinical trials. These models help optimize dosing, predict side effects and improve trial design.¹³

Challenges and Ethical Considerations

Despite its promise, AI in drug development faces several challenges. One major concern is data bias. Many AI models are trained on datasets that disproportionately represent certain populations, most commonly individuals of European descent and able-bodied males. This lack of diversity can lead to skewed results that fail to generalize across broader, more heterogeneous patient populations.¹⁴

In eye care, for example, this issue has already surfaced in the assessment of retinal nerve fiber layer thickness. Databases used by many leading optical coherence tomography (OCT) manufacturers are built on standardized datasets that predominantly feature white male subjects. As a result, diagnostic thresholds and normative values may not be accurate for patients from underrepresented ethnic groups, potentially leading to misdiagnosis or suboptimal treatment strategies.

AI systems also raise ethical and regulatory questions. Issues around data privacy, informed consent and algorithm transparency must be addressed. Many AI models function as “black boxes,” making it difficult to interpret how decisions are made. This lack of explainability can hinder regulatory approval and erode public trust.¹⁵

Furthermore, integrating AI into existing clinical workflows requires significant investment in infrastructure, training and cultural change. Resistance from stakeholders unfamiliar with AI technologies can slow adoption.

AI Assisted Drug Discovery: A New Era in Pharmaceutical Research

AI is not just a tool. It represents a paradigm shift in how we discover, test and deliver new medications. From identifying drug targets to optimizing clinical trials and personalizing treatments, AI has the potential to accelerate progress, reduce costs and improve patient outcomes.

As regulatory frameworks evolve and ethical standards mature, AI will become an indispensable part of pharmaceutical research. The future of medicine is intelligent, data-driven and increasingly collaborative. With deliberate stewardship and interdisciplinary innovation, AI will not only accelerate therapeutic breakthroughs but also reshape the very foundation of clinical research—bringing us closer to a more precise, inclusive and effective health care system.

References

PatentPC. The Cost of Drug Development. https://patentpc.com/blog/the-cost-of-drug-development-how-much-does-it-take-to-bring-a-drug-to-market-latest-data
PatentPC. Clinical Trial Success Rates. https://patentpc.com/blog/clinical-trial-success-rates-how-many-drugs-make-it-to-market-latest-approval-stats
Pharmaphorum. Oxurion Files for Bankruptcy After DME Drug Fails. https://pharmaphorum.com/news/oxurion-files-bankruptcy-dme-drug-fails
Chopra H, et al. Revolutionizing Clinical Trials: The Role of AI. Int J Surg. 2023;109:4211–4220. https://doi.org/10.1097/JS9.0000000000000705
Biomarker Research. AI in Target Identification. https://biomarkerres.biomedcentral.com/articles/10.1186/s40364-025-00758-2
PLOS Computational Biology. SAveRUNNER Framework. https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1008686
JAMIA. AI in Clinical Trial Design. https://academic.oup.com/jamia/article/31/11/2749/7755392
EPAM. ACUITY Platform Overview. https://www.epam.com/services/client-work/optimizing-data-for-smarter-clinical-trials
Digital ECMT. PROACT Patient Engagement Tool. https://digitalecmt.org/about-us/proact/
BenevolentAI. AstraZeneca Collaboration. https://www.benevolent.com/news-and-media/press-releases-and-in-media/astrazeneca-starts-artificial-intelligence-collaboration-accelerate-drug-discovery/
Saama. AI Clinical Analytics Platform. https://www.saama.com/news/saama-launches-industrys-first-ai-driven-data-platform-to-accelerate-clinical-development/
ACS JCIM. AI Drug Discovery Platforms. https://pubs.acs.org/doi/pdf/10.1021/acs.jcim.3c01619
VeriSIM Life. BIOiSIM Platform. https://www.verisimlife.com/our-platform
Popejoy AB, Fullerton SM. Genomics is Failing on Diversity. Nature. 2016;538:161–4.
Almeida D, et al. Ethics of AI in Healthcare. AI Ethics. 2022;2:377–87.

Author

Roya Attar, OD, DHA, MBA, FAAO , FORS

Roya Attar serves as an Associate Professor and Director of Optometric Services at the University of Mississippi Medical Center in Jackson, MS. As the sole optometrist at the state’s only medical school, she strives to make a meaningful impact on patient care, education and leadership. Dr. Attar has dual Doctorates in Optometry and Health Administration, complemented by a Master’s in Business Administration. A distinguished Fellow of the American Academy of Optometry, she assumes pivotal roles, including chairing the AAO Retina Special Interest Group and serving on the AOA Paraoptometric Resource Committee. In addition, she is president of her local society, where she contributes to discussions on best practices and professional development. Dr. Attar has been awarded: Mississippi Young OD of the Year, SECO Young OD of the Year, AOA Young OD of the Year, and Women in Optometry Young OD of the Year.

View all posts