A new exploratory survey from We Are Innovation finds that medical residents in Spain use artificial intelligence as a routine part of their work, including to support decisions about real patients. Most taught themselves, use free personal accounts, and have no hospital guidance to rely on.
Artificial intelligence is often described as the future of medicine. For a new generation of doctors in Spain, it is already the present.
Today We Are Innovation publishes The AI-Native Resident: How Spain’s newest doctors are using artificial intelligence. The study surveyed 30 medical residents (MIR) across 14 specialties about how they use AI, how they learned it, and what they need from the institutions around them.
AI is already part of the daily routine
87% of respondents use AI tools at least several times a week, and one in three uses them every day. Their use goes beyond studying. 97% have used AI to support decisions about real patients, most often when working through a differential diagnosis or checking treatment options. ChatGPT is the most widely used tool (90%), followed by the clinical search tool Vera Health (73%). 97% say AI saves them time, and 60% estimate they save at least an hour a week.
Adoption is happening from the bottom up
Residents have largely been left to work it out on their own:
- 90% learned to use AI by themselves, and only 7% have received any formal training.
- 93% say the MIR programme prepares them “little” or “not at all” to use AI properly.
- 80% access their main tool through a free personal account, and none reports institutional access.
- Not a single respondent knows of a hospital policy on clinical AI.
Residents are also clear-eyed about the risks. 97% worry about errors or “hallucinations.” Even so, only a third regularly check AI output against independent sources, which shows how much good training and guidance are needed.
What residents are asking for
Residents are not asking for AI to be held back. 83% believe health authorities should actively promote or facilitate its use, and the same share say they would use it more if they had adequate tools, training and clear guidelines. 83% want AI included in official MIR training. Nearly two-thirds worry that excessive regulation could limit access to useful tools, and the same share fear Europe is falling behind. Some clinical AI tools that doctors use elsewhere in the world are already unavailable in the EU.
“These discussions must take place, and our generation of young physicians must lead,” writes Dr Bernardino Azarías Ferrer Lozano, Health Futures Fellow at We Are Innovation, in the report’s preface.
“Adoption on this scale cannot depend indefinitely on individual judgment alone,” says Federico N. Fernández, CEO of We Are Innovation. “These residents want an enabling framework built on training within the MIR program, institutional access to reliable tools, and practical guidance.”
Six implications for policymakers
The report closes with We Are Innovation’s reading of the results, offered as a contribution to the debate:
- Invest in training, not restrictions.
- Offer simple guidance, not new layers of bureaucracy.
- Open the door to reliable tools.
- Trust professional judgment and strengthen it.
- Default to permissionless innovation, and keep Europe competitive.
- Build a stronger evidence base.
As the report puts it, the question should not only be what could go wrong if doctors use AI, but also what is lost when they cannot.
About the study
This exploratory survey was conducted between 18 September and 15 October 2026 among 30 medical residents in Spain (16 women, 14 men) through an anonymous online questionnaire. The sample is non-probabilistic and self-selected. The results describe the participants’ views and should not be taken to represent all residents or physicians in Spain. We hope the study lays the groundwork for larger, representative research across Europe.
