HealthManagement, Volume 24/25 - Issue 6, 2025
AI and generative AI are transforming healthcare by improving clinical documentation, diagnostics, drug discovery and personalised treatment. These technologies enhance efficiency, reduce costs and address clinician burnout while accelerating medical research. Despite their potential, challenges like data privacy, cybersecurity risks and reliability issues remain. Careful implementation, human oversight and strong data governance are essential to maximise benefits while ensuring patient safety and regulatory compliance.
Key Points
- AI reduces clinician workload by automating documentation, saving up to three hours daily.
- It improves accuracy in imaging and disease detection, enhancing medical diagnostics.
- With AI, drug discovery can be accelerated, cutting development time from years to months.
- Predicting treatment effectiveness with AI improves personalised care.
- AI adoption requires strong governance to address data privacy and cybersecurity risks.
The healthcare industry is at a crucial turning point as artificial intelligence and generative AI technologies transform the delivery, documentation and advancement of care. Healthcare systems around the world are facing significant challenges, such as ageing populations, rising costs and clinician burnout. In this context, AI technologies present promising solutions to some of the most pressing issues in healthcare.
Recent estimates suggest that generative AI could unlock nearly €970 billion ($1 trillion) in potential improvements within the sector, fundamentally changing areas ranging from clinical documentation to drug discovery. Market forecasts indicate that enterprise spending on AI and generative AI solutions is expected to grow substantially through 2028, highlighting the technology's increasing importance in the transformation of healthcare.
Key Applications Revolutionising Healthcare
The impact of AI and generative AI in healthcare spans multiple domains, with several key applications already showing significant promise. One notable area is clinical documentation, where AI-powered tools are dramatically reducing the administrative burden on healthcare providers. For instance, ambient AI scribes can automatically generate clinical notes from conversations between patients and physicians,potentially saving doctors up to three hours a day on documentation tasks. The Permanente Medical Group recently reported that their AI documentation tool was used in over 300,000 patient interactions. Remarkably, 81% of patients noted that their physicians spent more time making eye contact rather than looking at computer screens.
In medical diagnosis and imaging, AI systems are demonstrating impressive capabilities. Recent breakthroughs by Mayo Clinic and Microsoft Research demonstrate the power of generative AI in radiology, with new models that can automatically generate reports and evaluate chest X-rays. These advancements aim to improve clinician workflow and enhance patient care through more efficient and comprehensive image analysis. Additionally, researchers atStanford University have developed AI systems that can accurately diagnose skin cancer, matching the performance of dermatologists. Other systems are also enhancing the detection of breast cancer and lung abnormalities in medical imaging.
The field ofdrug discovery and development has seen perhaps the most dramatic acceleration through AI applications. Johnson & Johnson is leveraging AI agents to revolutionise drug discovery through the optimisation of chemical synthesis. These AI agents determine the optimal timing for solvent switches, which is a critical process in drug crystallisation. According to Jim Swanson, J&J's Chief Information Officer, this automation significantly speeds up processes that previously required multiple manual iterations by scientists, while still ensuring careful human oversight to prevent errors or bias. Similarly, other companies like Insilico Medicine have leveraged AI to reduce the time from target discovery to the selection of preclinical candidates from years to just a few months, all while significantly reducing costs.
Patient care and clinical decision support have also been significantly transformed by AI-powered solutions. In January 2025, Mayo Clinic and Cerebras Systems announced a breakthrough genomic foundation model that can predict treatment effectiveness for rheumatoid arthritis with 87% accuracy. This same model achieves 96% accuracy in predicting cancer predisposition and 83% accuracy in predicting cardiovascular phenotypes, demonstrating the potential of AI to revolutionise personalised medicine. Additionally, virtual health assistants can now provide 24/7 support for basic health queries, while advanced algorithms are used to predict patient risks and recommend preventive interventions.
Benefits and Transformative Impact
The benefits of AI integration into healthcare are becoming increasingly evident. Healthcare organisations that implement AI solutions are reporting significant improvements in operational efficiency, with some achieving cost reductions of up to 30% in administrative tasks. Additionally, the quality of care has improved due to more accurate diagnoses, a decrease in medical errors and the development of more personalised treatment plans.
Clinician burnout is a significant concern in healthcare, and AI-powered tools are being developed to help address this issue by automating routine tasks. Research shows that physicians can spend as much as 62% of their time on administrative work. By utilising AI automation, this burden can be greatly reduced, enabling healthcare providers to dedicate more time to patient care.
The timelines for research and development have been significantly shortened due to the use of AI applications. Traditional drug development cycles, which typically last 10 to 15 years, could potentially be reduced by as much as 50% with AI-assisted processes. Additionally, costs may be lowered by up to 70%, as indicated by industry analysts.
Challenges and Considerations
While the potential of AI in healthcare is promising, there are significant challenges that must be addressed for widespread adoption. Data privacy and security are major concerns, especially due to the sensitive nature of healthcare information and strict regulatory requirements such as HIPAA. Healthcare organisations need to ensure strong protection of patient data while also making the most of AI capabilities.
As AI agents become more autonomous in healthcare settings, concerns about cybersecurity are growing. Gartner predicts that by 2028, 15% of daily business decisions will be made autonomously by AI agents, while 25% of enterprise breaches will be linked to the misuse of these agents. This highlights the critical need for strong security measures and human oversight in the implementation of AI in healthcare.
The accuracy and reliability of AI systems remain critical concerns. While AI models have shown impressive capabilities, they can sometimes produce incorrect or biased results, particularly when trained on limited or non-representative data sets. This "hallucination" problem in generative AI models highlights the need for careful human oversight and validation of AI-generated content. An example of this is Johnson & Johnson's systematic review process for the outputs of their AI agents.
Implementation barriers also persist, including technical infrastructure requirements, staff training needs and integration with existing systems. Healthcare organisations must carefully consider how to implement AI solutions while maintaining operational continuity and ensuring staff adoption.
Looking Ahead
The future of AI in healthcare is expected to involve the growing use of autonomous AI agents collaborating with human professionals. However, as shown by leading organisations like Johnson & Johnson and Mayo Clinic, achieving success will require a careful balance. It is essential to leverage AI's efficiency while ensuring consistent human oversight to prevent errors and protect patient safety.
Organisations should consider several key recommendations as they move forward:
- develop comprehensive AI strategies that align with organisational goals and capabilities;
- invest in robust data infrastructure and security measures;
- ensure proper staff training and effective change management;
- maintain strong human oversight of AI systems;
- collaborate with experienced technology providers and ensure compliance with regulations.
As healthcare organisations gain more experience with AI technologies, we can expect to see increasingly sophisticated applications. The convergence of different AI technologies – ranging from natural language processing to computer vision – will likely lead to the development of even more powerful solutions for healthcare.
Conclusion
AI and generative AI represent transformative technologies for healthcare, offering solutions to long-standing challenges in efficiency, quality and access to care. While significant challenges remain, particularly around data privacy, accuracy and implementation, the potential benefits are too substantial to ignore. Achieving success will require careful planning, robust governance frameworks and a commitment to maintaining human oversight while leveraging AI's capabilities. As these technologies continue to evolve, healthcare organisations that thoughtfully integrate AI solutions while addressing key challenges will be best positioned to deliver improved patient care in the coming years.
Conflict of Interest
None