By Faye Miller
How can transdisciplinary researchers work with artificial intelligence as a genuine collaborator while maintaining integrative thinking? What new capabilities should be developed to ensure that artificial intelligence enhances, rather than fragments or compromises, cross-disciplinary human insights?
What can trandisciplinary researchers learn from human–artificial intelligence collaboration across disciplines?
Before moving on to capacities, let’s examine the growth in human–artificial intelligence research partnerships to see what lessons can be adapted by transdisciplinary researchers in their work. In particular, I suggest that integrative methodologies can be developed by understanding what is happening within the following domain-specific approaches.
Scientific research: The lengthy processes of hypothesis generation and pattern recognition can be accelerated by artificial intelligence, while human researchers can provide domain expertise and interpretive frameworks. For example, medication discovery teams use artificial intelligence to identify promising molecular compounds more quickly, then human researchers evaluate these findings using their expertise in biological, clinical, and safety contexts. This model suggests that preserving human judgement for connecting insights across disciplinary boundaries—a core strength of transdisciplinary work—is critical.
Complex problem-solving domains: Artificial intelligence processes massive datasets, while humans provide strategic thinking and stakeholder integration (including marginalised voices that may be sidelined by artificial intelligence systems). For example, social-ecology researchers use artificial intelligence to reveal patterns in environmental data, but human experts translate findings into policy recommendations that account for social, economic, and political factors. This trend boosts the transdisciplinary researcher’s role in synthesising technical findings with social implications and practical implementation.
Creative and content production partnerships: Artificial intelligence enhances initial research and idea generation, while humans steer with broader conceptual direction and contextual understanding. For example, writers collaborate with artificial intelligence for literature reviews, maintaining control over theoretical frameworks and narrative coherence. This practice suggests that artificial intelligence can rapidly analyse literature dispersed across multiple disciplines, while human oversight of theoretical integration and methodological coherence is preserved.
Five essential capacities for transdisciplinary researchers
The collaboration patterns described above reveal five capacities that transdisciplinary researchers could beneficially develop in order to leverage artificial intelligence effectively while maintaining their integrative approach.
- Recognising and evaluating how artificial intelligence systems perform across different disciplinary contexts
Transdisciplinary researchers would benefit from being able to go beyond technical understanding to recognise and evaluate how artificial intelligence systems perform across different disciplinary contexts. This means, for example, knowing when artificial intelligence excels, eg., at pattern recognition in quantitative data, versus when it struggles, eg,. with qualitative insights or cultural context. The development of expert judgement about which aspects of research benefit from artificial intelligence assistance, which do not benefit, and which require human integration, is essential for transdisciplinary researchers and is particularly important when working with datasets that span multiple disciplines with different methodological standards.
- Validating artificial intelligence-assisted findings
Transdisciplinary researchers would benefit from being able to validate artificial intelligence-assisted findings. This includes understanding how artificial intelligence biases might differentially affect various disciplinary components of the research and maintaining scientific rigour by developing protocols for human oversight.
- Creating workflows that smoothly integrate artificial intelligence and human inputs
Transdisciplinary researchers would benefit from being able to create research processes where artificial intelligence handles discipline-specific analysis, while humans focus on cross-domain synthesis and theoretical integration. For example, artificial intelligence analyses economic data, environmental indicators, and social surveys separately, and human expertise recognises connections and/or contradictions across domains.
- Providing stakeholder communication and translation
Transdisciplinary researchers would benefit from being able to explain artificial intelligence-assisted findings to diverse audiences—from disciplinary specialists to community stakeholders—while maintaining transparency about the role of artificial intelligence in generating insights. New forms of research communication could be developed to acknowledge artificial intelligence contributions without undermining human expertise and judgement.
- Providing ethical integration across domains
Transdisciplinary researchers would benefit from being able to acknowledge new challenges that arise when artificial intelligence is used in research where different disciplines and stakeholders have different ethical standards, ensuring artificial intelligence use remains appropriate across all domains involved. This capacity requires understanding how artificial intelligence decisions might have different ethical implications in different disciplinary and stakeholder contexts.
Orchestrating human–artificial intelligence collaboration
When researchers can fluidly orchestrate human–artificial intelligence collaborative contributions across different stages of their work while maintaining integrative oversight, then effective artificial intelligence-assisted transdisciplinary research can emerge. Not only does this require understanding what artificial intelligence and human expertise each contribute, but also paying attention to sequencing and combination processes.
Orchestration requires new forms of research design that explicitly account for artificial intelligence capabilities and limitations across different disciplinary and stakeholder contexts. It requires new research questions that leverage the pattern recognition of artificial intelligence while ensuring human oversight of theoretical integration. It also requires new approaches to validation that test not just individual artificial intelligence analyses, but the coherence and quality of insights that emerge from their integration.
For transdisciplinary researchers, human–artificial intelligence collaborations represent new opportunities to handle the complexity of transdisciplinary research more effectively, while maintaining the integrative thinking that defines their approach. Researchers who can thrive in this era are those who can use artificial intelligence to enhance their capacity for synthesis rather than replacing it. The future of transdisciplinary research does not depend on choosing between human and artificial intelligence, but on learning to orchestrate human–artificial intelligence collaborations in service of insights and breakthroughs that neither could achieve alone.
What have been your experiences with human–artificial intelligence collaborations? Are there institutional realities that enable or constrain how you are able to collaborate with artificial intelligence? How should we train current and future researchers to work well with artificial intelligence while continuing to educate them to think and work across different fields?
Use of Generative Artificial Intelligence (AI) Statement: Generative artificial intelligence was used in the development of this i2Insights contribution for initial framework scoping and research. (For i2Insights policy on generative artificial intelligence please see https://i2insights.org/contributing-to-i2insights/guidelines-for-authors/#artificial-intelligence.)
Biography: Faye Miller PhD is a research director, knowledge broker, author, editor, and career educator. She is Founder and Principal Consultant at Human Constellation Consulting, where she collaborates with technology companies, universities and scientific research organisations globally, on projects related to social and ethical aspects of science and technology, artificial intelligence literacy and ethics, information experience and shared understanding in transdisciplinary arts and sciences. She is currently based in Canberra, Australia.