IDC Guest Post
The life sciences and healthcare (LSH) industry sees immense potential in what AI offers and is investing heavily in AI(1) (NL2) Deep Pharma Intelligence reports: Cumulative investments have been made in AI-related drug development since 2014. In 2023, it will exceed $60 billion. AI, and GenAI in particular, is gaining traction in these industries, with one-fifth of LSH industries having conducted 7-10 internal and external product launches for GenAI use cases in the past year. (IDC Survey: Future Enterprise Resiliency & Spending Survey, Wave 4, April 2024). Roche has partnered with Silver Brain AI to develop IUCLID Assist, a GenAI tool that automates IUCLID6 submissions to the European Chemicals Agency by chemical and pharmaceutical companies. AstraZeneca, Merck, Amgen, and Sanofi are using GenAI for drug discovery.
While these industries have rapidly adopted public AI solutions to rapidly deploy their own AI solutions, the rapidly increasing number of cyber-attacks against the LSH industry has raised significant concerns. The chairman and CEO of a healthcare technology company called these attacks on infrastructure and personally identifiable information (PII) an “act of war.” Case in point: 52% of life sciences and 28% of healthcare cite the ability to provide robust data security as the most important characteristic when choosing a GenAI software provider (IDC GenAI ARC Survey, 2023 (August).
Compliance is critical in these highly regulated industries. Reduce the risk of privacy breaches and comply with data protection regulations such as the EU's General Data Protection Regulation (GDPR), the US Health Insurance Portability and Accountability Act (HIPAA), and the California Consumer Privacy Act. Essential (CCPA). Not surprisingly, compliance audits and PII detection were the two most important criteria for evaluating AI platforms in the LSH industry. (IDC Survey: Future Enterprise Resilience and Spending Survey Wave 4, IDC, April 2024). According to IDC's GenAI ARC study (August 2023), the two industries most affected by the GenAI revolution are life sciences, closely followed by healthcare. According to IDC's Life Sciences GenAI Survey (November 2023), in R&D, nearly 90% of industries are focusing their GenAI efforts on quality, risk, and compliance, and half of industries are focusing their GenAI efforts on creating. We are focused on optimizing medicines and clinical trials.
These regulated industries with sensitive data need alternatives to extend AI in a secure and compliant manner. Civilian AI is the answer. According to IDC's August 2023 GenAI ARC study (n = 599), 83% of information technology (IT) leaders believe that using GenAI models powered by their business data gives them a significant advantage over their competitors. I think you can get it. Private AI supports data encryption, differential privacy, data anonymization and pseudonymization, secure model governance, secure multiparty computing, and private cloud or on-premises deployment. This ensures data privacy and security, compliance with national and global regulations, better control over data and models, and greater customization and collaboration.
Benefits of private AI
Ensuring compliance with data privacy and security regulations. GDPR requires companies to respect patients' “right to be forgotten.” This means that if a patient withdraws consent to the storage and use of their data, companies must delete the patient's data. When using public AI, a third party may hold patient data and may not be able to honor patient requests to delete their data. Civilian AI can play an important role here. With private AI, your corporate data remains yours and you are responsible. Data encryption increases the security of your data. Privacy is further enhanced by differential privacy techniques that add noise to data to protect individual identities while allowing AI models to discover useful patterns. Protect your intellectual property: Because you use your own data to train your models, you can: Optimize your own models without sharing data with AI providers. AI providers will use the data to optimize public models that everyone can access. This becomes even more important when considering the intellectual property managed by the life sciences industry in areas such as drug discovery, design of experiments (DoE), cell and gene therapy manufacturing and supply chain processes. “Your data is our currency” is not an acceptable model for the life sciences industry, where data represents highly valuable intellectual property. By using controlled datasets and models, hallucinations and risks to patients can be minimized. Ensuring compliance with data sovereignty regulations: Countries around the world have implemented data sovereignty regulations that require protected health information (PHI) data collected or stored in a specific location to comply with local laws and be protected against violations. If you do, serious penalties will apply. Private AI allows you to control data flow. Improved collaboration: Access to a secure model gallery enables data scientists and application developers to securely collaborate across a portfolio of models. These models are scanned and validated before publishing to the gallery, and are privately stored and controlled by corporate IT, providing a robust and secure model collaboration tool with strong role-based access controls to support innovation. accelerate.
Investment in private AI for highly data-sensitive initiatives accelerates innovation and improves clinical outcomes
Public AI will continue to be widely leveraged in the LSH industry for more transactional AI initiatives that improve operational efficiency and accelerate the implementation of AI solutions. However, we are seeing an increasing shift towards developing private AI models that leverage enterprise data and develop solutions specifically for enterprises. Additionally, an IDC study on private AI commissioned by VMware found that one-quarter of the LSH industry prefers private AI models developed from scratch by their organizations due to multiple benefits, including: I found that I prefer using . Gain model versatility, greater model transparency, greater model accuracy, and greater cost efficiency customized to industry-specific use cases. In fact, 34% of the LSH industry believes that developing or deploying an AI model in the public cloud is just as expensive, and 28% believe it is more expensive than developing or deploying an AI model on-premises. (Source: VMware Private AI Survey, IDC, July 2024). These benefits are driving enterprises to expand their adoption of private AI solutions. For example, Moderna has developed 750 GPTs with hundreds of use cases driving positive value across the team.
IDC believes the LHS industry will invest in enterprise-wide adoption of private AI solutions (3) (NL4) to protect intellectual property and ensure data privacy. We believe this trend will become even stronger in fields such as precision medicine in the life sciences and medical image processing in medicine. . Under the Inflation Control Act (IRA), negotiated pricing applies after 13 years for biologics, compared to nine years for small molecules. Therefore, the focus of the life sciences industry will be on biologics. GenAI is transforming the discovery of new biologics, and the landscape of partnerships between innovative TechBios and major pharmaceutical companies is exploding. Private AI will be essential to protecting this sensitive intellectual property.
Ensuring data privacy is essential, and private AI is leading the way.
