AI literacy in healthcare: prompting is now a must

Dutch healthcare faces a historic shift: alongside hundreds of thousands of healthcare professionals, countless AI assistants such as ChatGPT, Copilot and specialised medical AI systems are this year active in day-to-day care. With the EU AI Act taking effect on 2 February 2025, this development gains a new dimension. While many healthcare organisations are still wrestling with ethical questions or limiting AI to basic administrative tasks, they risk missing a crucial development: from early 2025, concrete requirements apply to the AI literacy of professionals who work with AI systems. Instructing AI tools effectively (“prompting”) thus becomes not only a practical necessity but also a legal obligation for the modern healthcare professional, no less fundamental than making diagnoses, drawing up treatment plans or talking with patients. The AI Act applies across the EU, so this concerns hospitals and care providers in Belgium and Luxembourg too.

Two healthcare professionals working on prompts, generated with AI

The new dimension of medical expertise

This development requires more than superficial computer skills. It calls for a considered combination of medical expertise, systematic thinking and AI awareness. Healthcare organisations that invest now in solid AI training teach their professionals to write structured prompts. Take a physician working through an anonymised case, who asks an AI tool to set out: “1) a differential diagnosis based on the latest guidelines for chronic fatigue, 2) relevant biomarkers according to the updated 2024 laboratory protocol, 3) psychosocial factors according to the biopsychosocial model, and 4) possible drug interactions based on current prescribing policy.” A structured prompt like this produces a usable first analysis that still needs medical validation, but can support the diagnostic process considerably. AI literacy in healthcare therefore means that prompting is becoming both a practical necessity and a legal obligation.

Practical applications and organisational responsibility

The EU AI Act requires organisations to ensure, “to their best extent”, a sufficient level of AI literacy among staff who use AI systems. For healthcare organisations this is a concrete mandate to develop AI training programmes, introduce quality checks and draw up ethical guidelines. These requirements apply to all AI systems, with extra attention needed for systems classified as “high risk” in a medical context.

When writing medical reports, a well-formulated prompt can serve as the starting point for a first draft that brings in relevant guidelines and scientific insights. For medical research, healthcare professionals can instruct AI tools to analyse large volumes of scientific literature for specific developments or treatment protocols. In medication monitoring, AI can help identify interactions and contraindications. These applications do not replace clinical judgement, but they enrich the care process in ways that were previously unthinkable. Patient data stays subject to the GDPR throughout, so which tool you use for what matters.

Reorganising healthcare practice

Bringing AI into modern healthcare, combined with the requirements of the EU AI Act, calls for a fundamental redesign of care processes. Healthcare organisations need to redesign their workflows around AI-supported processes, with patient safety, quality checks and ethical considerations built in. That means investing in:

• Ongoing AI literacy training and education for all healthcare professionals
• Best practices for different medical specialisms
• Quality checks for AI-supported decision-making
• Drawing up and enforcing ethical guidelines for AI use in healthcare
• A culture that encourages sharing knowledge about effective medical prompting strategies
• Regular evaluation of AI competences within the organisation

Two doctors preparing for the new era, also generated with AI

A new era for healthcare

The urgency for the healthcare sector to embrace this development can hardly be overstated. Developments in AI systems follow each other at high speed, while the law sets concrete requirements for AI literacy. Organisations that do not act now risk both an unbridgeable gap and possible sanctions under the EU AI Act.

The future of healthcare lies in integrating AI effectively and responsibly as a complement to medical expertise, not as a replacement for it. The healthcare professional of the future needs to be not only medically skilled, but also AI-capable: able to tackle complex care questions with a combination of human expertise and AI support. This new reality calls for professionals who understand both the technical and ethical sides of AI and can apply that knowledge effectively in their daily practice.

Healthcare organisations that invest now in AI literacy and prompting skills are building not only compliance with the EU AI Act, but also their ability to deliver high-quality, innovative care in a changing healthcare landscape. In modern healthcare, more than ever: those who prompt, stay.