Can AI Repurpose Old Drugs for New Cures? The Unexpected Breakthroughs of 2026

The pharmaceutical industry has long grappled with an undeniable and often heartbreaking reality: bringing a new drug to market typically costs over $2 billion and can take 10 to 15 years, with a staggering 90% of compounds entering clinical trials failing to reach patients. This isn't just a financial burden; it's a human one, delaying life-saving treatments for millions. But my research into 2025 and 2026 data reveals something truly transformative: Artificial Intelligence (AI) isn't just optimizing drug discovery; it's fundamentally rewriting the rules, demonstrating the potential to slash development timelines by as much as 70% and reduce costs by up to 40% in critical phases. I've been tracking the integration of AI into healthcare, and what's emerging in drug development is nothing short of revolutionary. We're moving beyond theoretical models to tangible, measurable gains that are accelerating the pipeline from target identification to clinical trials. This isn't a distant promise; it's happening now, with profound implications for patients, pharmaceutical companies, and even the future of medicine itself.

The AI Engine Behind Drug Repurposing

One of the most compelling applications I've observed is AI-powered drug repurposing, where existing approved or investigational drugs are identified for new therapeutic indications. This approach is a strategic necessity rather than an experimental curiosity, driven by the sheer economics of de novo drug discovery, which averages $1 to $2.6 billion and 10 to 15 years per approved compound. AI dramatically reduces these barriers by leveraging machine learning, network biology, knowledge graphs, and transcriptomic signature matching to predict novel drug-disease associations. The beauty here is that these drugs already have established safety profiles, significantly cutting down regulatory risk and shortening the path to clinical trials.

I found that the COVID-19 pandemic acted as a forcing function, validating AI's power in repurposing when it was urgently needed to identify existing drugs active against SARS-CoV-2. Companies like Healx are specifically focusing on drug repurposing to accelerate treatments for rare conditions, demonstrating success in advancing therapies for patient populations with high unmet needs. This focus brings hope to millions affected by conditions previously deemed too niche for significant investment. AI can uncover the therapeutic potential of existing drugs for treating orphan diseases by integrating diverse data, including protein-protein interactions, multi-omics profiles, disease-specific molecular pathways, and clinical records.

Accelerating Novel Molecule Design

Beyond repurposing, AI is also proving instrumental in the de novo design of entirely new molecules. Generative AI models, including Generative Adversarial Networks (GANs) and deep reinforcement learning, are enabling researchers to identify novel drug candidates, optimize molecular designs, and predict compound efficacy with unprecedented accuracy. These models can analyze vast datasets of chemical and biological information, generating actionable insights that significantly expedite the traditionally labor-intensive discovery process. Companies like Novartis, for instance, are using generative AI to computationally design millions of potential compounds and create predictive models to assess key properties like brain penetration. This allows them to work with a handful of promising molecules in the lab, rather than synthesizing thousands, dramatically reducing the time needed to narrow in on a candidate.

Leading AI-driven drug discovery platforms are identifying viable drug candidates in months instead of years. Insilico Medicine brought its AI-discovered drug for idiopathic pulmonary fibrosis (IPF) from target identification to Phase II clinical trials in under 30 months, a process that traditionally takes 6 to 8 years and costs $100–200 million. This AI-designed drug reached Phase IIa at an approximate cost of $6 million. As of April 2026, no AI-discovered drug has received full FDA approval, but Insilico Medicine's Rentosertib (INS018_055) is the most advanced, having completed Phase IIa with clinically meaningful results, with the first approval projected for 2026–2027.

Streamlining Clinical Trials: A New Era

The impact of AI extends significantly into clinical development, traditionally the most expensive and time-consuming phase. AI is rapidly transforming how clinical trials are designed, conducted, and monitored, bringing automation, predictive insights, and real-time decision support into every phase of research. My analysis shows that AI can reduce clinical trial costs by up to 40% and accelerate trial timelines by 30-50% through predictive analytics for trial outcomes, protocol optimization, and enhanced safety monitoring. The global market for AI in clinical trials, valued at $2 billion in 2024, is projected to increase from $2.4 billion in 2025 to $6.5 billion by the end of 2030, growing at a compound annual growth rate (CAGR) of 22.6%.

One of the most critical bottlenecks, patient recruitment, is being addressed by AI-driven patient matching and outreach tools, which can cut recruitment timelines by as much as half. Furthermore, AI-powered predictive site selection models can lead to an average three-month acceleration in cumulative enrollment timelines, enabling trials to achieve up to a 37% faster enrollment rate. The IQVIA Institute's Global R&D Trends 2026 report found that AI-enabled programs at emerging biopharma companies showed a 75% Phase I success rate, a substantial advantage over non-AI programs. Recent industry data from early 2026 suggests that drug candidates designed via generative AI are achieving a 90% success rate in Phase I safety trials, nearly doubling the historical industry average of approximately 50%. This efficiency suggests a fundamental shift where drug development is treated more as a precise engineering challenge than a trial-and-error screening process.

Beyond Profit: Addressing Neglected Diseases

The economic implications of AI's integration into drug discovery are vast. McKinsey Global Institute projects that AI solutions could generate $60-$110 billion annually in value for the pharmaceutical industry, largely by accelerating early discovery and optimizing resource allocation. This translates not only to increased profitability for pharmaceutical companies but, crucially, to the potential for more affordable drugs due to reduced R&D costs. However, I've also observed a profound, unexpected angle: AI is making it economically viable to pursue treatments for rare diseases and neglected conditions, which traditionally received inadequate research funding due to their limited commercial viability.

With AI, the ability to repurpose existing drugs or rapidly design novel molecules with optimized properties means that even small patient populations can be targeted more efficiently. The FDA's N-of-1 pathway expansion in 2026, which allows individualized therapies for ultra-rare conditions, directly complements AI's capabilities, enabling single-patient evidence to meet regulatory standards when supported by robust genomic and biomarker data. Innovative trial designs that harness data augmentation, real-world data, and AI-enabled digital twins are enabling a new era of rare disease research, helping explore trial scenarios, and informing dose selection and study design.

What to Watch

I believe the biggest indicator of AI's truly transformative power in drug discovery will be the Phase III clinical trial results expected throughout 2026 and into 2027. These results will provide the first large-scale test of whether AI improves clinical success rates beyond the pharmaceutical industry's persistent ~90% failure rate. While AI excels at early discovery, its full impact on the most expensive and time-consuming part of drug development—large-scale clinical trials—is still evolving, facing challenges in regulatory adaptation and the fundamental complexity of human biology. Keep an eye on companies like Insilico Medicine, Recursion Pharmaceuticals, and Exscientia, who are at the forefront of bringing AI-designed drugs through clinical development.

Bottom line: AI is not a magic wand, and human biology remains complex. However, it is fundamentally changing the economics and timelines of drug development, making it faster, cheaper, and more accessible to address a wider range of diseases, including those previously neglected. The industry is moving from exploration to execution, with AI becoming a foundational part of the R&D operating model.