Skip to content
inklaretaal
No. 202641

Healthcare edition

in·klare·taal
Edition 202641 · Tuesday 6 October 2026 · Clarity since 2026

AI is growing in healthcare, but human judgement and training are lagging behind.

What this means for you
A model that learns from examples: training

Healthcare professionals use AI, but training is lacking

Dutch healthcare professionals are saving time with AI. But they want human oversight and better training.

AI is growing quickly in Dutch healthcare. A study shows where the benefits lie and what people are concerned about. It included 204 healthcare professionals and 2,009 patients. 63 per cent say their work is more efficient. 44 per cent spend less time on routine tasks and paperwork. At the same time, 83 per cent say AI training is missing or limited. 84 per cent think a person should check all AI output. Patients are also bringing AI information to appointments. Three-quarters of healthcare professionals say this happens. 68 per cent have had to correct incorrect AI information. So saving time does not automatically lead to careful use. Training and checks are still needed. The figures are based on people reporting their own experiences. Healthcare professionals and patients described what they experience. This is not an independent measure of time saved. The figures also do not prove that every AI tool is safe or useful. They do not say which tool saves time. So use the findings as a reason to agree on guidelines and training. They are not a reason to reduce human oversight. This week, ask your team what AI training is available. Also discuss which AI output needs to be checked in your work.

Source (in Dutch) →

Background for subscribers
What this means for you
First check who reviews AI output and who trains you to use it

AI saves time, but careful use does not happen automatically. This week, ask your team what AI training is available. Also agree who checks AI output before you act on it. This applies to record-keeping, alerts and answers from the admin team. Do not just ask a supplier for promises. Ask what the system is designed to do and what risks are known. And ask what is actually set out in a contract or law. Much of the news comes from abroad or has not yet been put into practice. So use it as a reason to agree on guidelines, not as a reason to change anything now.

Healthcare organisations get help to introduce AI

The Informatieberaad Zorg gave a positive recommendation on a plan for AI in healthcare. VWS is working on it with ten healthcare umbrella organisations. The plan has two aims. It helps organisations prepare. It also aims to make wider use of tools with a big impact. There will be self-assessments and a scan. These show how ready an organisation is, for example in terms of policy and roles. There will also be help with AI policy and with sharing responsibilities. The plan also mentions help with scaling up. Scaling up means introducing something that works in one place in more places. According to the source, the first support will not arrive until early 2027. So the positive recommendation does not mean teams will get help now. Nor does it mean a tool is immediately suitable. The source does not yet say exactly how the scan will work. This is mainly news for teams preparing policy or implementation. Ask your manager who is leading the preparations in your organisation. And ask which issues are still open.

Source (in Dutch) →

Background for subscribers

This is a taster.

Subscribers read the whole Healthcare edition: every story, each with its background in plain language.

Subscribe →

Not ready for that? Read the free letter first →

inklaretaal

AI and tech, made simple.

This is the Healthcare edition. Subscribers get the background to every story, in plain language.
Subscribe
inklaretaal · Amsterdam · © 2026