Walk into almost any clinic today and two kinds of work are happening at once. There is the human work: listening, explaining, and reassuring someone before a procedure. There is also the digital work: updating records, conducting video visits, and responding to alerts that appear mid-conversation.
Both matter. Yet many healthcare professionals received formal training in the first and learned the second through trial and error.
What has changed and why the job feels different
Care is more team-based than it used to be, and more of it happens through digital systems. That shift affects communication, record-keeping, clinical decisions, and what good practice looks like during a busy shift. It also changes what patients bring with them, since a large minority of adults now discuss their mental health with a general-purpose chatbot before they ever reach a clinician. Understanding what patients using chatbots have already been told, and what those tools cannot offer, is becoming part of the consultation itself.
Regulation is also changing. The European Union’s AI Act entered into force in August 2024, with its requirements taking effect in stages. In the US, the ONC HTI-1 rule makes USCDI v3 the baseline data standard for certified health IT from January 2026.
You do not need to memorise every rule or date. The important direction is clear: healthcare organisations need greater transparency about the tools they use, more consistent data, and a named person who remains accountable for decisions.
The people skills you need on every shift
AHRQ’s TeamSTEPPS programme treats communication as one of four essential skills supporting safe, efficient, patient-centred care. In practice, that means using plain words, keeping explanations focused, and checking that the message was understood. Teach-back is a simple approach: ask the patient to explain the plan in their own words.
When communicating with colleagues, a consistent structure reduces omissions. SBAR, which stands for situation, background, assessment, and recommendation, helps keep the essential information intact during a rushed handover or escalation.
Digital and data fluency
The AAMC organises its telehealth competencies into six areas: patient safety, access and equity, communication, data collection and assessment, technology, and ethical and legal considerations. The practical version is straightforward. Test the connection, check the sound and lighting, confirm the patient’s location and who else is present, and record what you could and could not assess remotely. Behind that setup, a UL TPU Cable is a neutral example of flexible, UL-listed cabling used in device assemblies and mobile carts.
A common data standard such as USCDI v3 helps information travel more consistently between systems and clinicians. It is therefore useful to know which fields use a set format, which allow free text, and how incomplete or vague entries could affect the next person providing care.
The equipment behind the screen
Digital care still depends on physical equipment, including monitors, mobile carts, sensors, leads, and connectors. Flexible, UL-listed TPU cabling is one example of this less visible layer. Its thermoplastic polyurethane jacket can provide the flexibility needed for certain device assemblies and equipment trolleys while meeting a recognised UL style.
Most clinicians will never specify such a component, but they still need to recognise equipment problems. A frayed lead, unreliable probe, or cart that will not charge is a fault to report, not a problem to work around. Learn who maintains equipment on your unit, how to log an issue, and which backup process applies when a device is unavailable.
The AI literacy you actually need
What to understand without writing code
Most clinical AI tools suggest, flag, draft, predict, or summarise. Their output can influence a decision, but professional accountability remains with the clinician. You should understand what a tool was designed to do, the information it uses, where it may fail, and when its recommendation should be questioned or overridden.
Greater transparency can make that task easier. HTI-1 introduced disclosure requirements for predictive algorithms used in certified electronic health record technology in the United States. FDA guidance also explains how manufacturers should document planned updates to certain AI-enabled medical devices. The practical lesson is that these tools can change over time, so staff need current training and clear local guidance. For readers comparing course options, artificial intelligence training can be a starting point; check for ethics, governance, clinical scenarios, and human oversight.
Use AI safely
The WHO’s guidance on large multimodal models, systems that can process more than one type of information, highlights risks such as inaccurate and biased outputs. In the UK, the GMC has also clarified that reporting duties for device-related incidents can cover software and AI tools. Before using a tool, confirm that it is approved by your organisation and that you understand its limits.
Where to build these skills
Short courses are one option, but course quality varies. Look for coverage of ethics, governance, realistic clinical scenarios, limitations, and human oversight. Resources from regulators and professional bodies are also useful. Start by asking your supervisor which tools your service already uses and what training is required for them.
How medical education is evolving
Curricula are gradually responding to these changes. The AAMC has published telehealth competencies and has been developing guidance for education in AI and other emerging areas. Its New and Emerging Areas in Medicine series covers subjects such as AI, quality improvement, patient safety, and telehealth. These materials are intended to support curriculum development rather than act as high-stakes accreditation standards.
Training remains uneven. A 2025 brief from the WHO European Observatory found gaps in digital skills training across the EU. It identified important areas including data literacy, communication, safety and security, problem-solving, and digital content creation.
Education structures differ between countries, so official sources are usually more reliable than broad summaries. For example, someone researching a medical school Hong Kong programme can use the HKU Li Ka Shing Faculty of Medicine website to find official information about undergraduate and postgraduate study, research, departments, and student support.
A 90-day learning plan
First 30 days: Choose one people skill and one technology habit
Use SBAR for every escalation. Alongside it, complete a short check before each remote consultation: connection, audio, privacy, patient location, and a fallback number if the call drops.
Next 30 days: Add a data or AI habit
Ask your supervisor to explain one algorithm or decision-support tool used in your service. Keep brief notes on what it helped with, where its limits appeared, and any occasions when clinical judgement led to a different conclusion.
Final 30 days: Share what you learned
Run a 20-minute team discussion on one communication, telehealth, or AI safety point. Afterwards, note whether the discussion changed a workflow or patient interaction.
Takeaway
People skills come first because they support every consultation, whether it takes place in person or remotely. Digital and data fluency matter because much of modern healthcare now depends on connected systems. AI literacy adds another layer: understanding what a tool does, recognising what it cannot do, and knowing when to question its output.
You do not need to develop every skill at once. Choose one practical habit, repeat it until it becomes routine, and then add the next.
FAQs
- What AI skills do clinicians need today? Clinicians should know what a tool was designed to do, what information it uses, how much confidence to place in its output, and when to escalate a decision. Documentation also matters because the reasoning behind a decision should remain clear when software has contributed to it.
- How can I practise remote consultation skills on an in-person placement? Rehearse a video consultation with a colleague and exchange feedback. Practise explaining a treatment plan without a physical examination, stating what cannot be assessed remotely, and confirming understanding aloud. These habits also strengthen face-to-face communication.
- Do I need programming skills to work with AI tools? No. Coding is a specialist skill rather than a general clinical requirement. Basic data literacy, curiosity about how a tool produces its output, and the confidence to question an unsuitable recommendation are more widely useful.
- What should I remember about patient privacy? Check that each tool is approved for use by your organisation. Do not enter identifiable patient details into an unsanctioned system, and consult your information governance team when the rules are unclear. It is safer to ask before sharing information than to address a privacy incident afterwards.










