‘Together the voice of AI’
Stories from the AI-aided Knowledge Discovery Lab
A running start
It’s been a while since we last caught up with and , the founding minds behind ASReview. If you know how it all started, you’ll remember it wasn’t your typical beginning: it kicked off with Rens sprinting after a tall stranger in a university hallway. That tall guy? Jonathan. The mission? Find someone who could turn a good idea into an even better reality. What began as a scribble on the back of a student thesis turned out to be something far bigger than either of them could’ve imagined. —an open-source tool for AI-assisted systematic reviews—quickly filled a huge gap in the research world. And it didn’t stop at “filling the gap.” It exploded. ASReview is now installed on more than 700,000 machines worldwide. From humble beginnings, it has grown into one of the university’s most successful (and some might say most beautifully out-of-hand) projects. It’s now standing shoulder-to-shoulder with commercial software giants in the field of systematic reviewing. The difference? ASReview is open, transparent, reproducible—and built by a community of researchers, coders, students, and professionals who believe that AI should serve science, not the other way around.
Built by many, used by more
ASReview has never been just a tool. It’s a gathering of minds from across disciplines and sectors. Data scientists, developers, information specialists, students, researchers, (government) institutions; people from inside and far outside academia. What brings them together isn’t just code, it’s a shared ambition: to rethink how we discover knowledge.
That ambition doesn’t stay indoors. It travels. In talks, workshops, and real conversations around the globe. Last year, four members of the ASReview team packed their laptops and headed across the Atlantic to Brazil: and , both researchers, were joined by and , two of the student contributors. They travelled with a clear mission: and let others experience ASReview firsthand. This wasn’t a one-off. Over the years, Bruno has become something of a travelling ambassador for the project. From to to , he’s been on the ground sharing, presenting, and connecting with researchers from all corners of the globe. And it’s not just about flying out. Sometimes, the community comes together in one place. One of the most striking examples was the second edition of the , Italy. A wild idea tested in Utrecht that grew into something genuinely game-changing.
The screenathon experiment
It started as a curious “what if”: what if we could crowdsource the screening process for systematic reviews? Instead of one researcher spending weeks (or months) sifting through thousands of articles, what if we brought people together and got it done in a matter of days?
In May 2024, that idea became real. and organised the first official Screenathon in Utrecht, where over 25 researchers from 10 European countries joined forces. Over the course of two days, they screened more than 12,000 publications in places like the Dom Tower and the botanical gardens. Reviewing turned into something collaborative, even celebratory. And best of all: it worked. In fact, it worked so well that “crowdscreening” became something more than just a new feature. It reshaped how ASReview was used. Not by simply assigning the same titles to more people, but by rethinking the process entirely: ASReview intelligently distributes records across experts, trusts their judgement, and uses AI to guide the screening dynamically. While other tools rely on collaboration to double-check the same papers, ASReview speeds things up by dividing the work and trusting expertise. So when LAB 2.0 was in development, the decision was obvious: crowdscreening had to be at the heart of the platform.
ASReview LAB 2.0: developed in dialogue
By May 2025, . It wasn’t just a software update, it was a milestone that brought together the full power of the community. It included contributions from active users, developers, machine learning researchers, domain experts, and students. This was ASReview at its most collaborative.
As such, the interface got a sleek redesign from Gen Z contributors and , two students who not only use the tool but represent the next generation of researchers. Their involvement wasn’t just about aesthetics; it was about designing with the future in mind. “I find it really important to focus on simplicity,” Angie explains. “It’s clear, transparent, and people can navigate their way through the software easily.” By involving younger voices in the development process, ASReview made sure the tool doesn’t just meet the needs of today’s users, but also feels intuitive, accessible, and relevant to those who’ll be using it tomorrow.
Behind that clean interface, the engine was quietly rebuilt. , , and were in charge of taking on the task of rethinking how ASReview makes decisions. They introduced a smarter foundation, with new default models that learn patterns more effectively and deliver better results, faster. One major step forward: ASReview now supports multiple languages, thanks to the integration of a multilingual model. Whether you're screening in English, Spanish, or Swahili, the system can follow along. They also experimented with more advanced model setups. Heavier, more powerful ones that push the limits of what’s possible with machine learning in screening. It’s become a tool that’s not only more intelligent, but more inclusive, ready for a broader range of researchers, topics, and languages.
A major contribution came from : a large and openly available collection of systematic review datasets, covering topics from medicine to education to social science. By offering this wide range, SYNERGY allows researchers from all kinds of fields to train and test ASReview in ways that are relevant to their own work, making the tool more flexible and broadly applicable.
And the impact is real. During the COVID-19 pandemic, the Dutch National Institute for Public Health (RIVM) used ASReview to help screen thousands of studies at speed. Today, they continue to apply the tool in literature reviews on topics like behavioural science and infectious disease, showing how academic code can become practical infrastructure for public policy.
Somehow, they’re all here now
It’s not just researchers who’ve embraced ASReview. Information specialists are among the most frequent users—sometimes by surprise. As , an information specialist, puts it: “Well, I—I didn’t start advising researchers to use ASReview… but some researchers advised us.”
And then there are the students. Some join to write a thesis. Some work as student assistants. Many never quite leave. ASReview offers them the energy of a start-up, but with one big difference: it’s all about science. As , one of those student researchers, puts it: “Working with ASReview not only introduced me to the power of active learning in systematic reviews, but also deepened my passion for research and academic innovation.”
From Vici Grants to podcasts: the momentum builds
Since our last update, the ASReview team has hit some impressive milestones. Rens was awarded a Vici grant of €1.5 million to continue pushing the boundaries of AI in science. The team was nominated for the , one of the most prestigious recognitions for scientific excellence in the Netherlands.
And none of this is slowing down. They keep building, testing, travelling, and—most importantly—doing it together.
Over the years, ASReview has grown into more than just an open-source project. It’s become a living, learning community of people who care deeply about how we discover knowledge, each bringing their own background, discipline, and perspective. From information specialists to machine learning researchers, from bachelor students to public health analysts, they work side by side. And somehow, it works: across fields, across time zones, across levels of experience.
And truth be told, no article can really do justice to the people behind this project. That’s why we launched the ASReview podcast: Inside ELAS.
The Inside ELAS podcast introduces you to the people shaping ASReview from the inside out. In each episode, you’ll hear what motivates them, what questions keep them up at night, and what it’s really like to co-create a tool that keeps growing, surprising, and challenging us. You can tune in on or watch it on . Want to dive in deeper? Follow ASReview on , join the conversation on , or and see for yourself what the future of reviewing looks like.