The global artificial intelligence in drug repurposing market stood at USD 1.3 billion in 2025 and is forecast to climb to USD 7.7 billion by 2033, growing at a CAGR of 24.5% between 2026 and 2033. North America leads with 52.9% share, oncology is the top application at 36.7%, and software & platform tools account for 66.6% of spend.

Baricitinib was designed as a rheumatoid arthritis drug. Within months of the COVID-19 outbreak, BenevolentAI's platform flagged it as a candidate for the virus, and by late 2020 it had emergency FDA authorization for exactly that use. That single case did more to legitimize AI-driven drug repurposing than a decade of academic papers — it proved a known, already-safety-tested compound could be redeployed against a brand-new disease in months, not years. That proof point is now the engine behind a market moving at nearly 25% annual growth.

What Is Driving Growth in This Market?

Traditional drug discovery is simply too slow and too expensive to keep up with disease burden. A new molecule typically takes over a decade and costs more than USD 2 billion to bring to market, with the overwhelming majority of candidates failing somewhere along that path. Drug repurposing sidesteps the most expensive and riskiest phase — early-stage safety testing — because the compound in question has already cleared it, often years earlier, for a different condition. AI's role is to make that search efficient: instead of scientists manually cross-referencing biological pathways, machine learning models scan genomic, proteomic, and clinical datasets simultaneously to surface drug-disease matches no human team could realistically find by hand.

Rare disease patients have become an unlikely growth engine for this market. Roughly 300 million people globally live with a rare disease, according to Lancet data — nearly 6% of the world's population — and about 80% of those conditions are genetic, with 70% appearing in childhood. Because these patient populations are too small to justify a de novo drug development program, repurposing existing approved drugs is often the only economically viable path to treatment. AI is uniquely suited here: it can extract signal from the thin, scattered datasets that rare disease cohorts typically produce, something traditional statistical methods struggle with.

Computing power has finally caught up with biological complexity. Multi-omics data — genomics, proteomics, metabolomics — combined with real-world patient records and decades of biomedical literature, creates a dataset too vast for manual review but exactly the kind of pattern-dense environment where deep learning thrives. Partnerships like Plex Research's April 2025 collaboration with Ginkgo Bioworks, which applied AI analysis to a large-scale transcriptomics dataset to uncover new disease mechanisms, illustrate how cloud infrastructure and improved algorithms are now making previously untenable analyses routine.

Download a free sample report or claim your copy of this full market intelligence report

How Big Is the Market, Really — and Where Is the Money Concentrated? 

Market size, 2025 - USD 1.3 billion

Market estimate, 2026 - USD 1.7 billion

Market forecast, 2033- USD 7.7 billion

CAGR, 2026–2033 - 24.5%

Leading region - North America (52.9% share)

Fastest-growing region - Asia Pacific

Leading component - Software & platforms (66.6% share)

Leading application - Oncology (36.7% share)

Software and platforms dominate spend today because that's where the actual scientific work happens — drug-target interaction mapping, biomarker discovery, and binding-site analysis all run through dedicated platforms like Innophore's CavitOmiX, launched with NVIDIA in January 2025. But services — data curation, model validation, regulatory-alignment support — are set to grow fastest, which tells you something structural about this market: platforms are becoming commoditized faster than the expertise needed to actually run them well. That's a classic sign of a technology moving from novelty to infrastructure.

Where Is AI Drug Repurposing Actually Being Used?

Oncology leads every other therapeutic area, and not by a small margin — largely because cancer research already generates the richest, most standardized datasets in medicine: molecular profiles, treatment-response records, and genomic sequencing at scale. AI models mine these datasets to reveal drug combinations that produce synergistic effects nobody had hypothesized to test.

Infectious disease repurposing is the segment to watch going forward. It's forecast to post the fastest growth rate through 2033, driven by AI's ability to screen existing drug libraries against host-pathogen interaction models — essentially asking "which of the 20,000 drugs we've already approved might also fight this pathogen?" This matters enormously for antimicrobial resistance, where the traditional new-antibiotic pipeline has been drying up for two decades; repurposing offers a faster, cheaper alternative when speed against resistant strains is the whole point.

On the technology side, machine learning and deep learning still do the heavy lifting, holding 45.8% share by powering the large-scale pattern recognition that underlies most repurposing predictions. But generative AI and large language models are the fastest-growing technology category, and this shift deserves more attention than it's getting. Earlier-generation ML tools were built to find correlations in existing data. LLMs and generative models can now read the entire corpus of biomedical literature, clinical trial registries, and patent filings simultaneously — then generate novel hypotheses about drug-disease relationships that were never explicitly stated anywhere in the source material. That's a move from pattern-matching to hypothesis generation, and it's a meaningfully different capability.

Pharmaceutical and biotech companies remain the dominant buyers, holding 59.2% of end-use share, for a straightforward reason: repurposing lets them extract additional value from existing drug pipelines and shelved compounds without restarting the entire development clock. Academic and research institutes are the fastest-growing end-use segment, increasingly functioning as the innovation layer that feeds algorithm development back to industry.

The Insight Most Coverage Misses

Nearly every market report on this space treats AI drug repurposing as a cost-saving story — cheaper, faster than building new drugs. That framing understates what's actually happening. The real shift is that repurposing is becoming a discovery mechanism in its own right, not a fallback option. Look at the pattern in recent deals: Insilico Medicine's November 2025 licensing collaboration with Eli Lilly isn't Lilly settling for a repurposed drug because a new one is too expensive — it's Lilly using Insilico's platform to co-design and optimize novel compounds against Lilly-defined targets, with repurposing methodology as the engine. The technology built to search for old drugs' new uses is now powerful enough to help design new molecules from scratch. That convergence — repurposing infrastructure feeding directly into de novo discovery — is where the next phase of value creation in this market will actually happen, and it's largely invisible if you're only tracking the market by "drugs repurposed" as a headline metric.

Looking for more in-depth data focusing on specific segments or regions? Get this report customized with inclusion of custom data sets to suit your exact business needs

What Should Buyers and Investors Watch Next?

Three signals matter more than the headline CAGR. First, watch platform-to-services ratio shifts — as more pharma companies build in-house AI capability, the services segment's faster growth suggests a market where expertise, not software licenses, becomes the scarce resource. Second, watch generative AI adoption specifically within infectious disease and rare disease applications, since these are the categories where thin datasets most benefit from a model that can generate hypotheses rather than just detect correlations in what already exists. Third, watch regulatory guidance closely — the FDA's draft frameworks on AI in drug development are still risk-based and evolving, and how agencies handle model interpretability and bias monitoring will determine how fast repurposing candidates can move from AI-generated hypothesis to clinical validation.

The Bottom Line

Drug repurposing was once viewed as pharma's consolation prize — what you did when a new molecule wasn't worth the bet. AI has flipped that logic. By making it possible to systematically mine the world's approved-drug library against nearly any disease target, repurposing has become a primary discovery strategy, not a backup plan, and the capital is following that shift at a pace few other pharma-adjacent technology categories can match.

Explore our dedicated business services:

• Brainshare Consulting – End-to-end business consulting services including Opportunity assessment, GTM support, Competitive intelligence, and Consumer Analytics.
• Custom Research – Get a market intelligence report tailored to your specific requirements and aligned with your business goals.
• Consumer Insights – Capture real, evolving consumer sentiment and behavior to help you make data driven strategies.
• Horizon Databooks – Access the world’s largest portal of Market Reports & Statistics
• Investment Insights – Make investment decisions with data driven insights, powered by domain and technology
• Signal (Pricing Intelligence) - Commodity price intelligence to drives strategic advantage. 

About us:

Grand View Research, a market research and consulting company, provides syndicated research reports, customized research reports, and consulting services.  Grand View Research database is used by the world's renowned academic institutions and Fortune 500 companies to understand the global and regional business environment. Our database features thousands of statistics and in-depth analysis on 46 industries in 25 major countries worldwide.