AI, quantum technologies, and drug discovery: 7 key guardrails

This article is brought to you thanks to the collaboration of The European Sting with the World Economic Forum.

Author: Satwik Mishra, Vice President (Content), Centre for Trustworthy Technology, Helen Huang, Senior Research Associate, C4IR Centre for Trustworthy Technology, World Economic Forum

• Less than 10% of drug development attempts are successful under legacy drug discovery processes.
• The convergence of AI and quantum technologies offers a unique solution to optimize fundamental processes in drug discovery.
• Adopting AI and quantum technologies also introduces implementation and ethical challenges, which are addressed in the trustworthy framework developed by the Centre for Trustworthy Technology.

Traditional drug discovery processes fail to address current and future global healthcare needs. The cumbersome processes, from initial R&D to clinical trials, can span a decade and are more likely to fail drug approval. In fact, more than 90% of drug development attempts are unsuccessful.

Unfortunately, this often means that humans are left defenseless against new and progressing viruses and infections. The United Nations’ World Health Organization predicts 10 million human deaths by 2050 from antimicrobial resistance (AMR), which is the microbial resistance to existing drugs.

Advancements in artificial intelligence (AI) and quantum technologies, however, could change that. These emerging technologies are working together to enhance existing drug discovery systems.

Quantum AI: a new era for drug discovery?

Deep learning methods, including novel optimization techniques, are useful in identifying and validating molecular targets. Once molecular targets have been identified, AI models search for and optimize compound interactions, expanding the discovery of effective chemical compounds for the given molecular targets. Meanwhile, quantum technologies accelerate AI capabilities. Advanced Graphics Processing Units (GPUs) allow AI models to achieve quicker results from more extensive data sets through parallel processing.

The symbiotic relationship between AI and quantum technologies has paved the way for a novel analytic tool: quantum machine learning (QML). Unlike traditional machine learning, QML embraces larger datasets while generating faster and more accurate results. These techniques are particularly useful when faced with complex optimization challenges, like searching for appropriate molecular binding orientations or improving drug absorption.

A new frontier for drug discovery

The Centre for Trustworthy Technology offers a detailed account of AI and quantum technologies’ role in critical phases of the drug development process, from initial discovery to late-stage clinical trials. The Centre’s recently released paper, A New Frontier for Drug Discovery and Development: Artificial Intelligence and Quantum Technology, provides a realistic overview of the technologies’ limitations and the challenges that pharmaceutical companies will inevitably face as they implement novel techniques in legacy systems and address ethical discourse.

To manage this potentially transformative technology and make the most of its opportunities, the Centre for Trustworthy Technology has created a framework to promote the trust, efficacy and proper use of these emerging tools. The framework, described as VITAL PIECES, articulates a careful and reasoned approach to applying AI and quantum technology to drug discovery processes.

Making quantum AI drug discovery work for all

The framework emphasizes seven key areas that will ensure the most effective and responsible use of this emerging tool:

1. The importance for companies to create detailed, customized risk mitigation plans, prioritizing transparency, documentation and justified procedural variations.

2. Regular assessments of AI models, vigilant data management to prevent biases and a robust data strategy to maintain integrity in the drug development process.

3. Clear communication with regulators and a multidisciplinary approach in designing solutions are inherent trust-building mechanisms for robust integration of these technologies.

4. A focus on safety and adherence to guiding principles throughout the drug discovery cycle is paramount, advocating for a broader industry-wide adoption of these principles.

5. Ethical cooperation and sharing best practices during the early stages of AI and quantum integration are vital for responsible adoption.

6. Educating the entire workforce on ethical principles and risk mitigation strategies is a priority.

7. Moving forward with a shared vision that balances the benefits and risks of these technologies is crucial to maintain trust and ensure faster access to drugs.

AI and quantum technologies offer immense benefits for the pharmaceutical industry. Yet, the industry must approach the transition with care, ethical diligence and a comprehensive understanding of drug development’s inherent challenges and responsibilities. Aligning this technological integration with established principles and ethical guidelines is essential to harness its transformative power while upholding the trust and safety fundamental to the industry.