From automating patient records to improving treatment outcomes, data has fuelled an invaluable digital transformation in healthcare. And with Generative AI, data’s role in improving healthcare is about to expand. GenAI promises to make every aspect of patient care more effective and personalised. But in parallel with the excitement, there are a lot of questions on how to get started on this journey.
A recent McKinsey survey found that 70% of healthcare leaders have either implemented GenAI or are in the process of testing it. In South Africa, almost half of healthcare leaders (47%) are already investing in GenAI and a further 37% plan to invest in it in the next 12 months.
The biggest areas of opportunity are patient experience and overall operations efficiency. But benefits can go beyond clinical uses into areas that improve interactions with patients, like AI-assisted chat support and personalised messaging.
GenAI thrives on data, which is where healthcare professionals can obtain the most gains. The best way to obtain value from this technology is to scan areas containing a lot of unstructured data. This means information that doesn’t follow a fixed format, like emails, images, videos and doctors’ notes. GenAI can automate tasks associated with this data, like connecting doctors’ notes to emails and images related to the same patient. Here are my key considerations on the opportunity ahead and how to put a strategy in place to get started.
Define clear objectives
Start by outlining specific goals tied to GenAI adoption – business problems you’re trying to solve. These should align closely with strategic healthcare priorities like improving patient outcomes, streamlining processes and enhancing operational efficiency. Identify pain points where traditional methods fall short and consider how GenAI will bridge these gaps. Examples include personalised patient care, automated medical documentation and virtual health agents for patient inquiries.
McKinsey’s survey suggests that most healthcare providers are beginning their GenAI adoption in clinically adjacent areas – processes like documentation, resource management and patient outreach. By easing GenAI into operational workflows, organisations can gain foundational experience while building trust in the technology.
Develop the required data infrastructure
Data quality is a top priority to avoid bottlenecks that can lead to AI project failures. Healthcare data is often fragmented and stored across different systems, so you need to clean up your data before moving forward with any projects. The recommendation here is to build a robust data infrastructure to support GenAI systems. This includes data pipelines, secure storage and tools to integrate data from diverse sources such as patient records, imaging systems and clinical notes.
Deploy solutions that can grow based on the volume of your data – at the core, edge and cloud. This isn’t just about planning: the right infrastructure saves money from your budget and allows you to invest in priority projects.
Frame governance from day one
Organisations need clear rules to make sure teams are using AI in a responsible way. Establish a governance structure that includes guidelines about using patient data, regulatory compliance and ethical decision-making. Cross-departmental collaboration between IT, clinicians and legal teams is essential here.
Address compliance with healthcare and technology regulations such as HIPAA, GDPR and FDA guidelines. Prioritise ethical considerations like fairness, accuracy and patient consent if you’re training and deploying GenAI models. Collaborate with legal and compliance teams to create a risk management framework, making sure you’re aligned with healthcare standards.
Foster a culture of innovation and learning
Build a culture that embraces innovation and experimentation. Pair IT teams with clinical leaders to co-design solutions that support what’s possible from a technical perspective but benefit clinical practices. Provide resources, such as internal GenAI-enabled tools and support systems, to empower teams. Ensure that all stakeholders understand the ethical and operational boundaries of using AI in healthcare.
Break down silos between IT, clinical teams and data scientists. A simple and effective action is to start a training programme that educates healthcare professionals on how to use GenAI tools effectively in their workflows.
Measure impact and scale up
Start small, measure impact and expand adoption based on proven successes. For example, if GenAI reduces a significant percentage of administrative errors in one department, scaling that solution horizontally could give you enterprise-wide impact. Continuous evaluation will ensure that the solutions you adopt meet their original value propositions.
Track progress through measurable KPIs like efficiency gains, cost savings and patient satisfaction levels. Use feedback to refine data models, address gaps in workflows, and identify new opportunities for deployment. Once successful, you can expand to additional departments or use cases.
The next frontier of healthcare
It’s clear that for healthcare, GenAI offers far more than incremental improvements. From shrinking operational inefficiencies to helping clinicians in their practice, technology is once again at the centre of an improved healthcare experience.
If you’re a tech innovator ready to take the first step, here’s where to begin. Develop a GenAI strategy that includes governance processes and clear frameworks to anticipate and manage challenges. The leaders in healthcare today are the ones asking proactive questions like, “What will the patient experience look like five years from now – and how can AI power that vision?”
The healthcare landscape is rapidly evolving, and GenAI isn’t just an option for staying competitive. It’s the compass guiding you to the industry’s future.
