Copyright infringement not intended

Picture Courtesy: Down to Earth

Context:

By 2026, healthcare systems across the world are facing unprecedented pressure. Ageing populations, the rapid rise of chronic diseases, persistent staff shortages, and escalating costs have stretched hospitals and public health institutions to their limits. In this environment, Generative Artificial Intelligence has moved swiftly from experimental pilots to routine deployment, transforming how healthcare is delivered, documented, and managed.

Generative AI in healthcare:

Generative AI refers to systems capable of producing human-like text, images, and predictions by learning from vast datasets. In healthcare, these tools are now widely used to draft clinical notes, summarise patient records, assist in medical imaging, support diagnostic reasoning, and accelerate pharmaceutical research. What sets this phase apart from earlier digital health initiatives is the speed of adoption and the deep integration of AI into everyday clinical workflows.

Current Status of Generative AI in Healthcare:

• The generative AI in healthcare market was valued at around USD 3.3 billion in 2025 and is projected to reach about USD 4.7 billion in 2026. It is expected to grow at a 7 % CAGR through 2035.
• Broader AI in healthcare (which includes machine learning, computer vision, NLP, etc.) was estimated at over USD 39 billion in 2025 and is expected to exceed USD 56 billion in 2026. 

• Around 66 % of physicians reported using AI tools in clinical practice by 2026, a significant rise from 38 % in 2023. 

• AI tools such as ambient scribe systems are automating real-time charting and clinical documentation, reducing clinician workload and burnout. 

• AI systems are widely used for diagnostics, imaging analysis, and risk prediction, becoming trusted decision-support tools across specialties. 

• Up to 50 % of healthcare operations now use AI for workflows like scheduling, revenue cycle management, and automation.  

Significance of Generative AI in healthcare:

• Addressing healthcare workforce shortages: Generative AI helps mitigate the global shortage of healthcare workers by automating documentation, reporting, and routine administrative tasks, thereby reducing clinician workload by nearly 30–50 per cent and allowing healthcare professionals to devote more time to patient care. 

• Enhancing operational efficiency and cost reduction: By streamlining scheduling, billing, claims processing, and clinical documentation, Generative AI significantly reduces administrative inefficiencies and healthcare costs, which account for nearly one-fourth of total healthcare expenditure in many countries. 

• Strengthening clinical decision support: Generative AI acts as a reliable decision-support system by assisting in diagnostics, medical imaging interpretation, and risk prediction, improving accuracy and reducing diagnostic delays while ensuring that final clinical judgement remains with human practitioners. 

• Expanding access to healthcare services: AI-powered virtual assistants and chatbots improve access to healthcare by enabling round-the-clock patient engagement, appointment management, and follow-ups, particularly benefiting rural and underserved regions with limited healthcare infrastructure. 

• Accelerating drug discovery and medical research: Generative AI shortens early-stage drug discovery timelines by rapidly analysing biological data and identifying promising compounds, thereby reducing research costs and accelerating the development of new therapies. 

• Improving patient experience and engagement: Through faster, clearer, and more consistent communication, Generative AI enhances patient satisfaction and continuity of care, especially for non-critical interactions and routine health management. 

Challenges of Generative AI in Healthcare:

• Risk of hallucinations and patient safety: Generative AI systems are prone to producing confident but incorrect outputs, with studies published in 2024–25 showing that large language models can generate clinically inaccurate information in 10–20% of complex medical queries, posing serious risks if used without human oversight. 

• Bias and inequitable health outcomes: Evidence indicates that AI models trained on skewed datasets perform unevenly, with research showing diagnostic accuracy dropping by 15–35% for underrepresented populations, thereby reinforcing existing disparities related to gender, ethnicity, income, and geography. 

• Data privacy and security concerns: Healthcare data breaches are rising, with global reports indicating that over 130 million healthcare records were exposed worldwide in 2023–24, making AI systems that rely on large-scale patient data vulnerable to misuse, cyberattacks, and consent violations. 

• Lack of explainability and transparency: Most generative models function as black-box systems, and surveys reveal that nearly 60% of clinicians are hesitant to trust AI-generated recommendations because they cannot understand or explain how the conclusions are reached. 

• Regulatory and legal ambiguity: Regulatory frameworks lag behind technological advances, with over 70% of countries lacking dedicated AI-in-healthcare regulations, leading to uncertainty regarding liability, accountability, and certification when AI-assisted decisions cause harm. 

• Over-reliance and deskilling of clinicians: Medical education experts warn that excessive dependence on AI tools may erode diagnostic reasoning skills, with studies noting that junior clinicians using AI support show reduced independent decision-making in up to one-third of assessed cases. 

Way Forward:

• Establishing ethical and governance guidelines: To address concerns around bias, fairness, and explainability, policymakers should adopt structured governance frameworks that prioritise transparency, safety, and ethical use of AI, following checklist-based methodologies that guide organisations on safety, fairness, and regulatory compliance tailored to healthcare settings. 

• Expanding National Digital Health infrastructure: Strengthening and scaling national digital health programmes, such as India’s Ayushman Bharat Digital Mission (ABDM), will build interoperable, secure health data systems that support AI tools while safeguarding patient privacy, confidentiality, and data access through standards-based digital architectures.
• Promoting responsible ai research & development: Governments should establish Centres of Excellence (CoEs) for healthcare AI, such as the Translational AI for Networked Universal Healthcare (TANUH) Foundation at IISc Bengaluru, to develop, validate and deploy AI tools for early disease detection and management, ensuring clinical collaboration and responsible design. 

• Enhancing data protection and health data governance: Extending policies such as the National Health Data Management Policy to explicitly cover support for AI tools will protect health data, ensure compliance with data protection laws, and promote the creation of representative, high-quality Indian health datasets for fair AI training and use. 

• Fostering global and multilateral collaborations: Participation in initiatives like the Global Initiative on AI for Health (GI-AI4H) under WHO, ITU, and WIPO can align national AI strategies with global standards, facilitate knowledge sharing, and accelerate ethical, evidence-based AI adoption in public health systems worldwide. 

• Supporting public health AI deployment programmes: Governments should pilot and scale AI programmes that address priority health challenges, such as AI-based cancer screening or health chatbots for maternal health services, providing digital tools that complement human providers while monitoring outcomes and refining governance. 

• Investing in skill development and workforce readiness: Public policy must prioritise training for healthcare professionals on AI literacy and ethical use, as well as building interdisciplinary capacity among data scientists, clinicians, and regulators to manage AI tools effectively and mitigate misuse or over-reliance. 

Conclusion:

Generative AI can play a transformative role in healthcare only when guided by strong government oversight, ethical frameworks, and robust digital health infrastructure. By aligning innovation with regulation, capacity building, and patient-centric safeguards, governments can ensure that AI strengthens healthcare delivery while preserving safety, equity, and human judgement. 

Source: Down to Earth