By Milena Francischinelli + AI
Co-Science is a simple idea: make research easier to do locally. It focuses on practical collaboration, a co-working space of science and research — across disciplines and institutions — so teams can share knowledge, reduce duplicated effort, and move from questions to results with less friction.
This isn’t a promise to “solve everything.” It’s a working approach for projects where complexity is real and no single field (or lab) can handle the whole problem alone.
What Co-Science is (in practice)
Co-Science supports research teams that want to work in a more open and coordinated way. It emphasizes:
- Shared goals and clear roles
- Transparent decision-making (what was decided, why, and by whom)
- Reusable outputs (data, methods, documentation)
- Inclusive participation (so collaboration isn’t limited to a small set of well-funded groups)
Where AI fits
AI can help with the busywork and the scale:
- summarizing papers and notes
- comparing methods and results
- organizing datasets and metadata
- detecting patterns or gaps in literature
- drafting protocols, reports, and early prototypes
But AI doesn’t replace scientific judgment. In Co-Science, AI is used as support infrastructure, while humans remain responsible for:
- framing the research question
- checking assumptions and quality
- interpreting results
- handling ethics, risks, and trade-offs
The human contact still matters
Collaboration only works when people can think clearly together. Co-Science treats human creativity and context as essential—especially when:
- evidence is incomplete
- values and consequences are involved
- “best” solutions depend on local conditions
AI can speed up exploration. Humans decide what counts as meaningful, safe, and worth pursuing.
Breaking silos without forcing “everything” together
Interdisciplinary work is valuable, but it can also become messy. Co-Science encourages cross-field collaboration when it’s useful—especially for problems like climate adaptation, public health, education systems, and sustainable development.
The point is not to merge disciplines. It’s to build clean interfaces between them: shared definitions, common metrics, and workflows that reduce misunderstandings.
Open access and data sharing (pragmatic version)
Co-Science supports openness because it often improves speed and reliability. But “open” needs structure:
- clear licensing and attribution
- privacy and security protections
- basic data standards (so others can actually reuse the work)
- documentation that explains what the dataset is—and what it isn’t
When done well, open sharing helps smaller labs and under-resourced institutions participate and contribute.