Building trust through independent scrutiny: Introducing the LeapSpace Advisory Board
Chair Jörg-Rüdiger Sack discusses the board’s ambitious plans to evaluate LeapSpace’s processes, content and policies and the importance of its external status.
Putting real people in the loop
A recent study found that nearly half of librarians and researchers feel more confident in AI tools “checked or produced” by recognized organizations, institutions or experts.1
LeapSpaceTM is developed in line with Elsevier’s Responsible AI Principles, which champion accountability through human oversight. That’s why real people run the regular evaluations of LeapSpace’s technology and responses and monitor the feedback channels. It’s also why we are recruiting a global independent panel — the LeapSpace Advisory Board — to evaluate and guide LeapSpace and its development.
Challenging assumptions from outside the “bubble”
The LeapSpace Advisory board is modeled on the successful Scopus Content Selection and Advisory Board (CSAB); a group of international scientists, researchers and librarians that vets and selects the content indexed in Scopus. It also regularly reevaluates that content, discontinuing titles that don’t comply with its guidelines.
Like the CSAB, the LeapSpace Advisory Board will be chaired by Professor Jörg-Rüdiger Sack, whose goal is to make it just as “rigorous, challenging and engaged”. More importantly, it will mirror the CSAB’s positioning — external and independent — something Jörg-Rüdiger considers essential.
“Once you're inside the bubble, you have a particular view of the world. You need some disruptive members who don’t have a vested interest to provide validation and credibility. It’s that stimulus that makes a product good. We’ve seen that with Scopus.”
Chancellor's Professor Emeritus Jörg-Rüdiger Sack, Professor of Computer Science, Carleton University
Today, Scopus is recognized as one of the most trusted and comprehensive abstract and citation databases. It’s why LeapSpace only uses abstracts and full-text journal content that are Scopus-indexed.
Meet the LeapSpace Advisory Board
Chancellor’s Professor Emeritus Jörg-Rüdiger Sack (Chair), Professor of Computer Science, Carleton University, Ottawa, Canada
Professor Holger Hoos, AI Centre RWTH Aachen University, Germany, and Leiden University, The Netherlands
Professor Jodi Schneider, University of Wisconsin-Madison, United States
Dr. Mikiko Tanifuji, National Institute of Informatics, Tokyo, Japan
Professor Fei Wu, Zhejiang University, Hangzhou, China
But while Jörg-Rüdiger Sack is passionate about board independence, he acknowledges that its members can’t succeed alone. “We need to work collaboratively with Elsevier. With the CSAB, that relationship is incredibly good, built on mutual trust and open conversation. That’s something you don’t often see on other boards. My goal is to recreate that with LeapSpace.”
Jörg-Rüdiger not only brings a wealth of board leadership experience, but the relevant subject knowledge — he is Chancellor's Professor Emeritus at Carleton University’s School of Computer Science in Canada. And he is working with Elsevier to recruit members who bring complementary skills in areas such as AI ethics and technology, research integrity, publishing and bibliometrics, and research infrastructure. “We’ve identified different categories of expertise. Clearly, not every board member will tick all the boxes, but together we will cover them,” he explains. “We also want members from diverse geographic regions.”
I think a board like this dedicated to an AI tool is pretty unique. I have certainly not seen or heard anything like this happening with other solutions out there.
Wim Meester, PhD
Head of Product Management, Scopus at Elsevier
Establishing the right checks and balances
Although recruitment will continue through 2026, board meetings are already underway with the five founding members, who are currently mapping out the board’s responsibilities. These will include:
Providing independent guidance on the algorithms used to surface and evaluate content within LeapSpace.
Working toward publisher-neutral content discovery and search ranking.
Advising on responsible AI use in academic and scientific research settings.
Supporting transparent and explainable AI outputs for researchers.
As Jörg-Rüdiger notes: “Trust in AI depends on more than how powerful the technology is. Researchers need confidence that the underlying content is selected fairly, that algorithms operate transparently and that the same standards are applied consistently across publishers.”
In practice, this will involve zeroing in on specific areas of LeapSpace’s development and decision making. “One of the first questions we will ask is whether the AI models incorporated in LeapSpace are the right ones. Are they sufficient to achieve the short- and long-term objectives? And are they prone to hallucinations and bias?” Jörg-Rüdiger explains. “Datasets are also important. We will look at whether those being used are reliable, trustable and certifiable in some way. And whether any are missing.”
The board will also evaluate the methodologies used to test LeapSpace and check to see whether LeapSpace is addressing the issues that researchers care about – and ensuring that intellectual property and privacy are respected. It will also consider the quality of LeapSpace responses, along with the contextual data points, or “trust indicators”, provided to help users evaluate them.
The importance of publisher neutrality
Alongside abstracts from more than 7,000 publishers, LeapSpace indexes a growing volume of peer-reviewed full-text content — currently more than 23+ million articles and books, comprising 42% of all global research articles published since 2021.2 This full-text content expansion is driven by the content licensing partnerships we are signing with other publishers and societies in pursuit of our goal to make LeapSpace a truly publisher-neutral solution, which is something Jörg-Rüdiger believes will be crucial to LeapSpace’s widespread adoption. “It’s critical that the scientific community does not view LeapSpace as an Elsevier-biased or Elsevier-only tool. To date, Elsevier has made great steps in signing these agreements with other publishers, but we will be scrutinizing their progress to ensure that pace continues.”
The board will also be closely involved in ensuring that the responses LeapSpace delivers are publisher neutral — a task that indicates the complexity of the work that lies ahead for the board.
For example, members must first agree a common definition of publisher neutrality. Jörg-Rüdiger explains: “You cannot just say it means all publishers should be equally represented. That wouldn’t work.”
Board members have suggested framing it as "equal treatment of equivalent evidence" across all evidence classes (supporting, contradictory, critical, emerging). And they have already agreed that neutrality must be observable and evidence-based, not simply a matter of stated intent.
However, questions remain around what form that evidence will take. Wim explains: “For us, publisher neutrality is that we apply the rules in LeapSpace equally to every publisher. But how will the board evaluate that process? We are working together to establish what information we can supply that will enable them to confidently say yes, LeapSpace is publisher neutral.”
Jörg-Rüdiger adds: “We also need to think about how that publisher neutrality is communicated to the user. For that, transparent policies will be important, so will need to evaluate those too.”
Researchers told us that one of their biggest concerns with AI tools is hidden bias. That’s why our objective is to certify LeapSpace’s ranking methodology to be completely publisher-neutral — no preferential treatment.
Professor Jörg-Rüdiger Sack
Chair of the Scopus Content Selection and Advisory Board | Subject Chair CSAB - Computer Science | Professor Emeritus at Canada’s Carleton University’s School of Computer Science
For Jörg-Rüdiger, a key measure of the board’s success will be user satisfaction; something he believes takes many forms, from time savings and ease of use through to pride in an AI-assisted output. On a personal note, he is also interested in ensuring that LeapSpace strikes the right balance between answers and inspiration. “I’m seeing something other educators have also observed — when my graduate students go to an AI tool with a problem, it takes them straight to the solution, and they miss that whole step of preparing a phase-by-phase project plan. This leaves them frustrated and unsure what to do next. Some may even give up. Sometimes it’s better to hold back some information to let people find their own way. I hear from the LeapSpace development team that they’ve built LeapSpace to encourage critical thought... So that’s something we’ll be checking too.”
References
1. Rapple, C. (June 2026). Taming the Crocodile study – phase 1 report. Kudos. http://info.growkudos.com/croc-pubopens in new tab/window. This report was sponsored by Elsevier and other leading publishers
2. Elsevier analysis using Scopus, June 2026.
Next steps
Interested in learning more about LeapSpace? Sign up for email updates to get the latest news and articles with best practices, user insights, and product updates.
