Melvine's AI Analysis # 44 -AI in the Lab: How Novartis is Rewriting the Rules of Drug Discovery
Melvine Manchau
Senior Strategy & Technology Executive | AI & Digital Transformation Leader | Former Salesforce Director | Driving Growth & Innovation in Financial Services | C-Suite Advisor | Product & Program Leadership
April 7, 2025
Artificial Intelligence (AI) and Generative AI (Gen AI) are rapidly reshaping industries globally, with healthcare and pharmaceuticals among the most profoundly transformed sectors. Novartis, a leading global pharmaceutical company, has emerged as a frontrunner in harnessing AI capabilities to drive innovation, enhance operational efficiency, and revolutionize patient care. This article explores Novartis’s AI and Gen AI initiatives, industry trends, competitor strategies, anticipated impacts, inherent risks, challenges, and the evolving regulatory landscape.
AI and Gen AI Initiatives at Novartis
Novartis has been proactively embedding AI and Gen AI into its core operations, from drug discovery and clinical trials to manufacturing and personalized patient care. Key initiatives include:
1. Accelerated Drug Discovery and Development
Novartis employs advanced AI algorithms to significantly shorten the drug discovery timeframe. AI platforms rapidly screen vast molecular datasets to identify promising drug candidates, predict biological effectiveness, and simulate clinical outcomes. Generative AI further enhances this process by designing novel molecules with optimized therapeutic efficacy, safety, and manufacturability.
The development of novel therapies and drugs - The use of AI is being explored in the pre-clinical phase to understand disease biology and drug candidates; in the clinical phase to help target populations and to design intervention studies; and in the development of digital therapeutics and devices to enable continuous monitoring. Generative Chemistry: The use of generative chemistry to augment chemistry teams with well-annotated, high-quality ideas in a seamless fashion for our end users. Novartis uses Machine Learning to scan billions of molecules in our compound library and propose virtual molecules with a desired target profile, as defined by our drug discovery experts. It efficiently reports the multiparametric ideation process every medicinal chemist undertakes daily. The output is a manageable set of optimized compound suggestions that can be readily synthesized. Discovery scientists can either directly choose from select compounds or be informed to come up with related, yet novel ideas.
MELLODDY7 (Machine Learning Ledger Orchestration for Drug Discovery) The MELLODDY project (Innovative Medicines Initiative consortium, of which they are part) has created an AI platform that learns from proprietary compound assay data (>one billion data points for 10 million small molecules) contributed by multiple pharmaceutical companies, while maintaining confidentiality through blockchain-based encryption. Companies retain control over their data and the resulting Machine Learning models. The models learn correlations between chemical substructures and activities in biological assays of disease relevance and benefit from techniques such as ‘transfer learning’, the principle that prediction accuracy may be enhanced by learning from models in adjacent areas. MELLODDY will enable cheaper, faster, and higher-throughput drug discovery by providing structure-activity information for legacy and current assays in our drug discovery pipeline.
Case Example:
Partnership with Microsoft: In 2019, Novartis launched an AI Innovation Lab with Microsoft to leverage AI for accelerated drug discovery and personalized medicine. Using Azure AI, Novartis researchers systematically explore complex molecular interactions, drastically reducing the time required for identifying viable drug candidates.
2. Clinical Trial Optimization
AI algorithms enhance clinical trial efficiency by optimizing patient recruitment, site selection, and data analysis. Gen AI techniques generate synthetic patient data to simulate clinical scenarios, enabling Novartis to design more robust and predictive clinical trials, thus accelerating regulatory approval pathways.
Use Case Highlight:
Sensyne Health Collaboration: Novartis partnered with Sensyne Health, leveraging machine learning to analyze anonymized patient data and identify optimal trial candidates, significantly enhancing clinical trial efficiency and reducing costs.
3. Precision Medicine and Patient Care
Novartis deploys AI to personalize treatment plans, enhance diagnosis accuracy, and predict patient outcomes. AI-driven predictive analytics identify patient populations most likely to benefit from specific therapies, particularly in oncology, cardiology, and rare diseases.
Notable Initiative:
AI-driven Precision Oncology: The company utilizes advanced analytics to determine genetic markers and biomarkers predictive of treatment responses in cancer patients, thus tailoring therapeutic strategies to individual patient profiles.
4. Manufacturing and Supply Chain Efficiency
AI and machine learning optimize Novartis’s manufacturing processes and supply chain management, improving operational efficiency, forecasting accuracy, and quality assurance. Predictive maintenance algorithms reduce downtime and maintain high compliance and safety standards.
Buying Engine: AI-powered marketplace for Novartis. Designed to streamline and centralize purchasing decisions, Buying Engine aims to enable procurement efficiency across Novartis by creating a ‘one-stop-shop’ algorithmic-based marketplace, starting with lab supplies, PPE, and potential spare parts (indirect material). This system aims to provide transparency and recommend optimal buying choices in near real-time, leveraging multiple techniques from knowledge representation, recommender systems, optimization, and Machine Learning algorithms to achieve its goal.
The optimization of business processes and operations
AI is being explored and may improve clinical development, manufacturing, and supply chain processes by automating, optimizing, and re-engineering processes. In the business services, we use AI to ensure efficiency, effectiveness, and drive operational excellence and compliance.
AE Brain: Automating repetitive processes. AE Brain improves the quality of our safety information and also reduces the burden of repetitive manual work. AE Brain processes messages to identify potential adverse events and technical complaints in these messages. The system ingests textual data from multiple sources and applies Natural Language Processing (NLP) technology to understand the contents of those text documents to identify adverse events. This system is integrated into the workflow of human experts as a decision support system. Marketing Mix Models Marketing mix models (MMMs) are statistical models for measuring the effectiveness of marketing activities such as promotion, media advertisement, etc. These models can be of many types, but multiple regression is the workhorse of most marketing mix modeling. Regression is based on a number of inputs (or independent variables) and how these relate to an outcome (or dependent variable) such as sales, profits, or both.
5-. Engagement with patients, healthcare professionals, and partners
AI is being explored to enhance engagement with stakeholders and participants in the healthcare systems to support patients and generate insights. For Novartis to sustainably improve and extend people’s lives, they must collaborate with trusted partners in tech, academia, and other areas.
Ai Nurse: Empowering Patients. Novartis partnered with Tencent to develop a WeChat mini-app called Ai Nurse for patients diagnosed with heart failure. The patient engagement platform empowers patients and their healthcare providers to be more aware of their condition and take appropriate actions to improve their health and well-being. The app uses multiple AI-driven algorithms to transform voice to text and text to voice. Algorithms are used to anticipate disease progression, recommend activitie,s and provide targeted coaching and education. This data is continuously assimilated and interpreted to assess a patient’s improvement or worsening condition. Accordingly, nurses and physicians can remotely track patients, with full consent and privacy protections as discussed in this paper, and provide additional continuity of care recommendations.
Trends in the Pharmaceutical Industry Regarding AI
The strategic integration of AI technologies increasingly characterizes the pharmaceutical sector. Prominent trends include:
Generative AI in Structural Biology: AI-driven protein-folding and structure prediction platforms (e.g., AlphaFold from DeepMind) are revolutionizing drug target identification.
Real-World Data Utilization: Expanding real-world evidence (RWE) powered by AI analytics to enhance clinical trial and drug approval decision-making.
AI-based Diagnostics and Digital Therapeutics: Growth in AI-enabled digital tools for early diagnostics, patient monitoring, and personalized digital interventions.
Collaborative AI Ecosystems: Increasing cross-sector partnerships between pharma companies, technology giants, and startups, driving accelerated innovation.
According to an article by https://www.genengnews.com/, Novartis and Generate: Biomedicines Sign Up-to-$1B AI Protein Drug Collaboration. The Companies will combine the Generate platform with the pharma giant’s expertise in target biology, biologics, and clinical development.
“The technology we’re working on is protein modality agnostic and protein disease area agnostic. So, it doesn’t matter whether it’s a neuroscience or a cardiovascular target, because the technology has applicability in those domains,” Nally added. “We just need the relevant expertise to develop the right hypothesis.”
That’s where Novartis is expected to help. In addition to its expertise in drug discovery and development, the pharmaceutical giant has years of experience in AI, stretching back to 2019, when it selected Microsoft as its strategic AI and data-science partner for establishing an “AI innovation lab” intended to “transform how medicines are discovered, developed, and commercialized.”
“It’s focused on what we call generative chemistry and AI-driven drug discovery, Novartis CEO Vasant (Vas) Narasimhan, MD, said of the Microsoft collaboration at a media event last year. “Our goal now is to invest even more as technology gets better and better within that space.”
“A lot of these natural language processing capabilities could allow us to accelerate and simplify many parts of R&D,” Narasimhan predicted.
In an article posted April 29 on the company’s website, Bülent Kızıltan, PhD, Novartis’ global head of AI & Computational Sciences, declared: “Our ultimate goal is to transform the entire drug discovery process from an AI-enabled to an AI-enhanced and, finally, to an AI-driven process, where the majority of the work will happen in silico.
Competitor Initiatives and Strategies
Novartis’s competitors are equally active in adopting AI solutions:
1. Pfizer
Pfizer employs AI-driven drug discovery platforms, partnering with companies like CytoReason and Insilico Medicine to advance immunology and oncology treatments.
2. Roche
Roche leverages AI extensively in diagnostics and personalized medicine, notably through its Flatiron Health subsidiary, applying AI to analyze large-scale oncological patient data and drive precision oncology.
3. AstraZeneca
AstraZeneca collaborates with BenevolentAI to harness generative and predictive AI models for drug discovery and repurposing, targeting previously challenging diseases.
Expected Impact of AI and Gen AI at Novartis
The integration of AI and Gen AI technologies at Novartis is expected to deliver substantial benefits, including:
Reduced Drug Development Timelines: Accelerating processes from discovery to market, significantly decreasing current timelines of 10–15 years to potentially 5–7 years.
Cost Efficiency and Operational Optimization: Increased productivity and reduced operational costs through predictive analytics and intelligent automation.
Enhanced Patient Outcomes: Precision medicine driven by AI analytics improves treatment efficacy, patient adherence, and long-term health outcomes.
Innovation Leadership: Securing industry-leading positions in novel treatment areas and fostering competitive differentiation.
Risks and Challenges of AI Implementation
Despite significant opportunities, several risks and challenges accompany AI adoption in pharma:
Data Privacy and Security: Handling sensitive patient information requires stringent compliance and robust cybersecurity frameworks.
Algorithmic Bias and Fairness: Potential biases in training data can lead to unequal treatment outcomes, posing ethical and reputational risks.
Validation and Explainability: AI-driven decisions demand transparent and explainable algorithms for regulatory approval and clinician trust, especially in clinical contexts.
Talent Acquisition and Skill Gaps: The limited availability of skilled professionals adept in AI and life sciences poses a challenge to sustained innovation.
Novartis is committed to using AI systems responsibly and in alignment with the commitments and principles articulated in their Code of Ethics: 1. Empower Humanity 2. Hold Ourselves Accountable 3. Mitigate Bias 4. Respect Privacy 5. Be Transparent and Explainable 6. Assure Safety and Security by Design 7. Prioritize Environmental Sustainability 8. Review, Learn, and Adapt
Regulatory Environment and Compliance Considerations
The regulatory landscape surrounding AI in pharmaceuticals is evolving rapidly, with authorities such as the FDA, EMA, and others developing specific guidelines:
FDA’s Digital Health Initiatives: The FDA’s Digital Health Center of Excellence promotes the adoption of AI-driven digital therapeutics and diagnostics and provides clearer regulatory frameworks.
EMA’s AI Guidelines: The European Medicines Agency (EMA) issued guidelines emphasizing transparency, accuracy, and robustness of AI algorithms used in clinical trials and diagnostics.
Data Protection Regulations: Stringent compliance requirements under GDPR (EU) and HIPAA (US) necessitate robust data governance frameworks within pharmaceutical companies.
The Future Outlook for Novartis and AI
Novartis’s proactive embrace of AI and Gen AI positions it strategically to lead transformative changes within the pharmaceutical sector. By leveraging AI’s potential to accelerate discovery, optimize operations, and personalize patient care, Novartis aims to drive significant improvements in clinical outcomes and operational efficiency. Nonetheless, navigating inherent regulation, ethics, and data management challenges requires sustained vigilance and responsible innovation.
As AI technologies mature and regulatory landscapes evolve, Novartis’s continued investment and strategic partnerships in AI and Gen AI will likely solidify its industry leadership and reshape global healthcare paradigms.
By Melvine Manchau, Digital & Business Strategy at Broadwalk and Tamarly
https://melvinmanchau.medium.com/
