During a working session at the AIET Forum, participants grappled with a fundamental tension: we're building tools to use AI in education, yet most academic journals prohibit AI use in submissions. The unjournal concept emerged as a radical rethinking of how research communities can disseminate knowledge faster and more accessibly.
The field changed since we showed up here a couple days ago. Traditional journals take a year to publish. Things are happening too fast.
Why an Unjournal?
AI research moves faster than traditional academic publishing. By the time a paper is reviewed and published, the field has already moved on.
Traditional academic journals were designed for a slower era of research. The typical pipeline—submission, peer review, revision, publication—can take 6-18 months. In a field where capabilities double every few months, this creates a fundamental mismatch.
The unjournal proposes a different model: rapid, open dissemination with community review happening in parallel rather than as a gatekeeper.
- Quick: Not waiting a year to get something out
- Accessible: Someone can read it and understand it
- Living: Content can evolve as the field evolves
Participant: "It's a new type of dissemination. That's quick. It's not waiting a year to get something out. No, because things are happening too fast. Like, the field changed since we showed up here a couple days ago, basically."
We're researching AI in education while traditional journals prohibit using AI. If we're creating an unjournal about AI, shouldn't we embrace AI in the process?
A provocative question emerged: if we're talking about AI in education and creating an unjournal, shouldn't we be exploring how AI can enhance the research process rather than banning it?
Potential uses discussed:
- Integration: AI helping to synthesize multiple contributions into coherent papers
- Linking: Creating networks between related content in the repository
- Summarization: Generating accessible summaries of complex discussions
- Exploration: Creating "AI lenses" for different ways to explore the repository
Participant: "Current journals actually keep you from using AI, so the question is, can the unjournal allow for using AI in certain ways so that there's a benefit for the reader or the authors? I mean we're talking about AI and then we're creating an unjournal which makes total sense, to me, and then if we would go the same route and be like, no AI in there, then that's a little weird, right?"
AI-Powered Discussion Summaries
A new workflow for capturing working sessions: AI transcribes and summarizes, experts review, community validates. "That's brilliant."
The breakthrough idea: use AI to capture and synthesize the rich discussions happening in working sessions—discussions that typically disappear after the meeting ends.
Participant: "Maybe creating a set of prompts, we will record it and AI will transcribe and based on the prompts come up with a good summary of what they discussed, what they agreed."
Another voice: "That's brilliant. That is brilliant. Then we don't need to worry about it, right? Even people who, you don't need to watch, you know, if you record it, you can watch everything and it's really hard to summarize, and then the discussing person can go through and make sure that the AI summary is accurate and can provide further feedback."
Knowledge Forum Inspiration
The Knowledge Forum model from Toronto: a collaborative infrastructure where people comment on and build upon each other's work. Now imagine it with AI enhancements.
Knowledge Forum, developed at the University of Toronto by Carl Bereiter and colleagues, has been a model for collaborative knowledge building since the 1980s. The system allows users to build on each other's ideas, creating a web of interconnected contributions.
The unjournal draws inspiration from this model:
- Commentary layers: Like Reddit, but for research—community comments on contributions
- Linked ideas: Cross-references between related content
- Tracing development: Network analysis showing how ideas evolve
- AI integration: Using AI to surface connections and patterns
Participant: "Knowledge Forum, right, from Toronto... Knowledge Forum is an infrastructure that people do this and they comment on each other's work. And, you know, it was a very sophisticated idea when it first came about in, I don't know, the 80s, really, right?"
Another voice: "It's like Twitter without the lunacy. It's like a sophisticated Twitter. They're commenting on each other."
As the repository grows, AI can create different "lenses" for exploring content—epistemic networks, keyword connections, and intelligent search.
One of the most exciting possibilities: using AI not just to create content, but to help users explore the growing repository in meaningful ways.
- Epistemic networks: Visualize connections between concepts discussed across papers
- Smart synthesis: AI can summarize entire collections, not just individual contributions
- Multiple perspectives: Different AI "lenses" for different exploration needs
- Quality commentary: AI summaries become another way of "commenting on the AI"—humans as expert validators
Participant: "Over time, if we have this repository that's growing with more stuff, you could almost create AI lenses for the community to use to explore the repository in different ways."
Another voice: "Especially in a repository, because usually, like, you have the instrument here, but then the article that you want to read here, and then you have a body of research that uses those things, and they're all in separate places, right? So, if this is all in this unified way, then your summary could be just about that whole collection."
A repository with video clips, instruments, commentary, AI-powered summaries, and epistemic network analysis—all in one unified, fast-moving, community-validated platform.