During our first AIRON AI Leadership Conversation, Professor Elena Gaura made an observation that deserves particular attention from those of us working in research management and research support. She suggested that professional research services may increasingly need to stay “three steps ahead” of the research community.
This is not about knowing more about science than the researchers we support. It is about understanding how research is changing, anticipating the implications and helping institutions prepare before new challenges become urgent.
As AI becomes more deeply involved in the research process itself, this ability to anticipate change may become one of the most important capabilities of our profession.
When AI starts contributing to research itself
Professor Gaura introduced the concept of auto-research: the increasing automation of parts of the process through which new knowledge is generated.
AI can already help researchers analyse large bodies of literature, identify connections between findings, develop hypotheses, write code and interpret data. Increasingly, these activities may become connected within more automated research workflows.
Imagine an AI-supported system that reviews thousands of publications, identifies previously overlooked relationships and proposes new research directions. Researchers would still need to determine which questions matter, evaluate the evidence and interpret the results, but the process of developing scientific knowledge could look very different.
AI could change not only how quickly research is conducted, but how research questions are developed, evidence is assembled and new knowledge is produced.
This creates exciting opportunities for discovery, but also raises questions that our existing research systems may not yet be equipped to answer.
New capabilities, new questions
Research funding and governance frameworks have largely been designed around established approaches to producing and assessing research.
But what happens when AI contributes substantially to generating hypotheses, developing methodologies or analysing results?
How should originality and intellectual contribution be assessed? Who is accountable when an AI-assisted analysis produces incorrect conclusions? How will peer review evolve when both applicants and reviewers increasingly use AI?
These questions reach beyond technology. They concern research quality, integrity, transparency and trust.
One of the important points raised by Professor Gaura was the potential gap between what researchers can now do with AI and what institutional policies and funding frameworks are prepared to address.
Funders may develop different expectations, and institutional guidance may struggle to keep pace with emerging practices. Researchers could find themselves using capabilities for which clear policies do not yet exist.
This is where research professionals have an important role to play.
Working between researchers, funders and institutional governance, we are well placed to recognise emerging tensions, anticipate new requirements and help develop responsible approaches.
Rather than waiting for new rules to appear, we can contribute to shaping how institutions respond.
Where should automation stop?
As AI becomes more capable, the question is not simply what can be automated, but what should be automated, under what conditions and with whose oversight.
AI can already assist research support teams with funding searches, eligibility checks, reporting, document preparation and portfolio analysis.
However, determining whether a research opportunity is strategically valuable, evaluating risk or making decisions about scientific significance and ethics requires context, expertise and accountability.
The value of AI is not in removing human judgement, but in enabling people to make better-informed decisions.
Research professionals will increasingly need to understand this distinction and help design workflows in which AI supports human expertise rather than replacing it.
From administration to strategic intelligence
This transformation also creates an opportunity to reconsider the role of research professionals.
Much of our work has traditionally centred on administering processes, interpreting funding requirements, managing compliance and coordinating research activities.
These responsibilities remain important, but AI could significantly reduce the time spent on repetitive and transactional tasks.
Imagine research managers spending less time assembling reports and more time interpreting research portfolios, identifying emerging risks, advising academic leaders and contributing to institutional research strategy.
The shift is not simply from manual work to automation. It is from processing information to using information more strategically.
Research professionals could increasingly become a source of institutional intelligence, connecting developments in research practice, technology, funding and policy to support better decisions.
This will require new capabilities, including an understanding of AI, research integrity, data governance and emerging funder expectations. Equally important will be the confidence to question established processes and contribute to discussions about how research support services should evolve.
What does staying three steps ahead mean in practice?
Staying three steps ahead does not mean predicting every technological development or having all the answers.
It means developing the ability to recognise change early and respond thoughtfully.
For research professionals, this could mean three things:
- Understand what is changing. Explore how researchers are already using AI, including emerging applications across the research lifecycle.
- Anticipate the implications. Identify where existing funding requirements, governance arrangements, research integrity standards or institutional processes may need to evolve.
- Help shape the response. Work with researchers, funders and institutional leaders to develop practical approaches that enable innovation while protecting research quality and accountability.
These challenges cannot be addressed by individual institutions alone. Professional communities and networks such as AIRON have an important role in enabling shared learning, bringing different perspectives together and helping our profession prepare collectively.
The opportunity and the responsibility
Auto-research could expand what researchers are able to investigate, helping them explore questions and connections that were previously beyond their available time and resources.
But greater technological capability does not automatically mean better research. Its value will depend on how we combine it with critical thinking, integrity, expertise and human accountability.
For research professionals, this presents an opportunity to move beyond supporting existing processes towards helping institutions rethink how research is developed, supported and assessed.
Our role may increasingly be not just to navigate the research system, but to help shape how that system evolves.
Perhaps this is the real meaning of staying three steps ahead: not having all the answers, but recognising which questions we need to be asking before everyone else starts asking them.
Leave a Reply