WASHINGTON: A groundbreaking legislative proposal in the United States seeks to redefine the role of artificial intelligence in healthcare by classifying AI as a "practitioner licensed by law" capable of prescribing FDA-approved drugs. The proposed amendment to section 503(b) of the Federal Food, Drug, and Cosmetic Act (FFDCA) would formally recognize AI and machine learning systems—subject to state statutes and FDA clearance (under sections 510(k), 513, 515, or 564)—as legal prescribers.
Proponents argue that leveraging AI’s advanced data analysis and pattern recognition capabilities is inevitable, especially in a healthcare landscape challenged by persistent doctor shortages. Notably, Bertalan Meskó, MD, PhD, a renowned expert in medical innovation, commented, "While it might sound like a dangerous step, using AI to even prescribe drugs is inevitable due to global doctor shortages. If AI can spot signs of cancer on a CT scan or pedestrians from a self-driving car's camera, in certain cases, it will be able to prescribe medications too. All these decisions are based on data and AI is incomparably better at analyzing that."
If enacted, this legislative change could enable AI systems to manage routine prescribing tasks, alleviating the burden on overextended medical professionals and potentially improving patient access to care. However, the proposal has ignited a robust debate among healthcare experts and industry leaders. Many question accountability, asking, "Who would be responsible if AI issues a wrong prescription?" This concern underscores the urgent need for clear liability frameworks to ensure patient safety.
Julio Bonis Sanz, MD, MBA, AI NLP Engineer, and Epidemiologist, emphasized the necessity of defining stringent criteria to distinguish advanced AI systems from longstanding rule-based algorithms—a distinction that has become increasingly blurred with the advent of large language models. Similarly, Manoj Bhojwani, a Pharma/Medical Device/Life Sciences Executive, suggested that regulatory easing for AI-driven diagnostics should precede its use in prescribing medications. Meanwhile, Samuel Hale, an advocate for AI-driven wellness, acknowledged AI’s potential to augment clinical decision-making but stressed that ultimate accountability must remain with human professionals.
Additional voices from the industry have raised concerns over potential biases in AI training data and the risk of exacerbating issues like overprescription. Experts from varied backgrounds argue that while AI excels in data analysis, healthcare decisions require the nuance of human judgment to address ethical dilemmas, ensure equitable care, and maintain patient trust.
Regulators and industry stakeholders are now calling for a balanced approach that fosters innovation while establishing robust safety and accountability measures. As debates continue, the potential impact of this legislation on healthcare delivery remains a topic of intense scrutiny—one that could fundamentally reshape the future of medical practice if implemented with the necessary checks and balances.