What Three UK Events Showed Me About the I4R Network
May 22, 2026 · Juan P. Aparicio
Last week, I4R felt less like an organization and more like a living research network. On Monday, researchers gathered in Cambridge. Two days later, another group met at University College London. The next day, we were at King’s College London. Different rooms, different participants, different local teams, but one shared purpose: to understand how AI is changing research.
For me, that was the real story behind the paper we are writing. The study asks how AI affects the way people carry out difficult tasks, and whether its benefits look different for junior and senior researchers. But the UK events also showed something way cooler: I4R has grown into the kind of network that can take a question like this and actually study it!
Across the three events, roughly 300 participant slots were randomized. Participants worked across 20 papers and were assigned either to an AI-assisted condition or to a human-only condition. They spent the day doing the kinds of things researchers actually do: reading unfamiliar work, navigating data and code, checking results, looking for possible mistakes, thinking through robustness, and writing referee reports.
The numbers matter, but they are not what I will remember most. What I will remember is the sight of so many people, across three institutions, taking this work seriously. Some participants were students. Others were experienced researchers. Some were very comfortable with AI tools. Others were not. But all of them were helping us study a question that has become unavoidable: how will AI change the way research is done?
This is where the I4R network matters. A study like this requires much more than a good idea. It requires coauthors who can design the study, local organizers who can make each event work, participants willing to give a full day of serious effort, and a shared infrastructure for turning all of that activity into evidence. No single researcher could do this alone. Probably no small team could either.
That is why I left Cambridge, UCL, and King’s feeling both exhausted and pleasantly surprised. The surprise was not that people care about research quality. I have seen that many times through I4R. The surprise was seeing how far the network has grown. We were able to coordinate three events in one week, across major institutions, around a common research design, with enough participants and collaborators to ask a genuinely important question.
For me, this is the most exciting part of the project. The paper gives us a framework for studying AI and expertise. But the network gives us the ability to do it properly. I4R is becoming a community that can move quickly, coordinate across places, and produce evidence on questions that matter for the future of science.
The paper we are writing is the frame. The network is the story. And after seeing Cambridge, University College London, and King’s College London come together in one week, I felt more convinced than ever that I4R can help answer some of the most important questions, not only about the effect of AI, but also about how research is done.