Skip to main content

1Cademy

Understand new science. Then question it.

AI is helping produce new science faster than experts can check it. 1Cademy connects small, sourced topics through learning prerequisites, so scientists can learn missing background and question each contribution.AI-assisted science now outpaces expert checking. Learn what a result builds on, then question each contribution.

5:37 · Narrated story

Explore the graph

From a new result to shared understanding

First story step.
Last story step.
Proposed shared-learning workflow. Recent AI research reports illustrate AI-assisted results; a Google DeepMind policy paper and Vannevar Bush's 1945 essay give context for the growing volume of new work. Published abaucin research provides a teaching example; the readers, graph and community revisions are illustrative.

Faster than experts can check.

Faster than experts can checkA generic new result appears first, produced with an AI's help. As the narration says AI is helping produce new science faster than experts can check it, more AI-assisted results land beside it as a growing pile that stands taller than the new result by the words can check it, while an expert reader stands beside it. Four attributed cases then appear one at a time: Google DeepMind, new algorithms; Microsoft, coolant prototype; Anthropic, toxin binders; OpenAI, math results. Each case's brief status arrives as the narration names its check: auto-evaluated as automated evaluators score the algorithms, tested in-house as researchers make and test the prototype, lab-tested binding as the labs confirm binders, and for examination as the proofs are released; shorter stages leave the statuses out. As the proofs are released, OpenAI's proof sheet fans out into several sheets, and a magnifier then moves over the proof for mathematicians to examine. A note says the logos imply no affiliation. These are company reports with different AI roles and levels of checking, not equivalent validated discoveries or institutional endorsements. Anthropic's card shows, in a molecular viewer, one designed binder docking onto a toxin and its contact lighting up, after other designs try the toxin and turn away where the stage has room, a schematic with no numbers; its record says Claude designed the binders, working mainly autonomously for about 64 hours, and two contract labs found that 17 of 211 designs bound, the strongest with single-digit nanomolar affinity; no neutralization or animal data is reported yet, and people gave prompts during the run. Where the stage has room, as the narration says leaders once kept up with their own fields, a senior scientist appears beside a short pile of results, each checked as it lands, and the reports' pictures dim while their names stay legible; as it says now even they cannot, the pile rises far past her head, new AI-assisted sheets landing unchecked faster than she can check one more. As the narration asks how many more scientists can learn to check such results, the pictures return, four more readers join her, two either side, each holding a magnifying glass toward her pile, and the closing label reads Learn, then check. On wide desktop stages she and her pile stand in the gap between the two columns of cards and the readers flank the closing label; where that gap is narrow, the pile rises through it from the strip along the bottom of the stage, where she and the readers stand, with the closing label to their right; on phones and portrait tablets they stand in the row below the reports, the pile rising between the report columns; on wide, short stages they stand between the two columns of reports; on landscape stages with room they stand below the row of reports, the pile rising between the middle two. Stages with no room below or between the reports, such as short phones, leave out the leader and the readers; the closing label still appears. The sources and qualifications are inspectable in the record.

1 / 10 · Faster than experts can check

Inspect record

Proposed shared-learning workflow. Recent AI research reports illustrate AI-assisted results; a Google DeepMind policy paper and Vannevar Bush's 1945 essay give context for the growing volume of new work. Published abaucin research provides a teaching example; the readers, graph and community revisions are illustrative. The abaucin source is the authors' abstract, not reconstructed study data. Clinical effectiveness and safety in people are not established here. Company names and logos identify each report's source and do not indicate endorsement.

This is a proposed workflow, not a claim that separate scientific topic and link review, records of which version was used, or objection records are deployed. Existing platform proposals and comments are not a scientific approval process. Learning and review benefits remain to be tested, including whether granular review reduces missed errors or bias.

Prerequisite arrows point from background to the topic that builds on it. The existing AI ranking to result arrow stays as it is. Separately, an illustrative human-proposed link from AI ranking to Tested spectrum is shown as pending; an AI reviewer objects, and people review the objection and decline the link. The concentration-test prerequisite stays as it was. When Mara traces backward, she moves against the arrows; the arrows themselves do not change.

An AI drafts a clarification, Mara and other readers ask for what it leaves out, the AI revises, and human reviewers adopt shared version 2. Topic content and prerequisite links receive separate decisions. Sharing is not scientific approval. Neither decision changes the paper, its data or its finding. Sources, earlier versions, attribution and the link objection are inspectable below. Supported alternatives can coexist with objections and dissent in the proposed workflow; no substantive alternative dispute is depicted here. Keeping an objection does not make it true or equally supported. Original records retain their original versions.

The assay graphics are schematics, not actual measurements, experimental protocols or a complete curriculum. Concentration means compound quantity per volume of test medium. Mara knows AI ranking; the second reader already knows both AI ranking and concentration tests. Familiarity is illustrative, not a mastery assessment. Learning helps frame questions, not prove a finding or replace specialist review.

Full illustrative sequence, including later decisions, not only the selected chapter.

Illustrative spectrum revision

Version 1: Earlier shared explanation
An activity spectrum compares responses across the bacteria tested.
Version 2: Shared after review
An activity spectrum describes responses across the bacteria and conditions tested. A narrow spectrum means a compound inhibited only a few of the bacteria tested. Its interpretation depends on test concentrations, assays and the tested panel; it is not a claim about every untested organism.
AI contributor
Clarify the scope of the tested spectrum.
Mara and other readers
Make test concentrations and conditions explicit.

Adopt the clarification as shared learning material, not a judgment that the scientific result is true.

Revision source: authors' abstract (opens in a new tab)

Separate link proposal: Declined

AI ranking to Tested spectrum. Proposed by Human contributor; objection by AI reviewer.

Understanding a laboratory activity spectrum does not require understanding AI candidate ranking.

Keep the existing concentration-test prerequisite; do not add the proposed AI-ranking link.

Bacterial growth
Bacterial growth is the increase in the number of bacterial cells as they divide; a growth assay measures it under specified conditions. A growth assay does not by itself establish a treatment's effect in a person.
Concentration test
In a laboratory assay, concentration means the quantity of a compound per volume of test medium. An antibacterial test relates concentration to measured growth under stated conditions. Growth inhibition is not the same claim as killing every bacterial cell.
Tested spectrum
An activity spectrum describes responses across the bacteria and conditions tested. A narrow spectrum means a compound inhibited only a few of the bacteria tested. Its interpretation depends on test concentrations, assays and the tested panel; it is not a claim about every untested organism.
AI ranking
A trained model can prioritize compounds for experiments. Model ranking and experimental evidence are distinct parts of the published research.
AI-assisted result
Liu and colleagues reported AI-assisted identification of abaucin with narrow-spectrum activity against Acinetobacter baumannii and investigated it in a mouse wound model. These findings do not establish clinical effectiveness or safety in people.
Liu et al.: Deep learning-guided discovery of an antibiotic targeting Acinetobacter baumannii (2023) (opens in a new tab)

The authors' abstract supplies the real AI-assisted research context. The illustration does not reproduce the study's data or depict a 1Cademy evaluation of the compound.

OpenAI: Working with academic communities to build scientific understanding (2026) (opens in a new tab)

Context for the broader task of explaining and examining AI-generated research, not an endorsement or evaluation of 1Cademy.

Google DeepMind: AlphaEvolve: A coding agent for scientific and algorithmic discovery (2025) (opens in a new tab)

The authors report new algorithms from an evolutionary coding agent with automated evaluators. The paper describes an autonomous pipeline; people supply each task, a starting program and evaluation code. This is the developers' own white paper, not an independent evaluation of all results.

Microsoft: Transforming R&D with agentic AI (2025) (opens in a new tab)

Microsoft reports AI- and simulation-assisted identification followed by synthesis of a non-PFAS coolant prototype. Its May 2025 announcement said some primary properties aligned with predictions, the prototype was under further analysis, and the project was an experiment. The report does not establish a mature material or wholly autonomous laboratory work.

Anthropic: How Claude designed protein binders against scorpion neurotoxins (2026) (opens in a new tab)

Anthropic's researchers chose five scorpion toxins; Claude then picked their structures and binding hotspots and directed 12 protein-design tools to create binders, working mainly autonomously for about 64 hours. Two contract labs, Anthropic's partners, tested 211 designs and 17 bound their target toxins. Of the 60 designs each aimed at a single toxin, 11 bound, a rate the authors call consistent with typical protein-design campaigns; the strongest binders reached single-digit nanomolar affinity. The report includes no neutralization or animal data yet, so it does not show an antivenom or a treatment. People gave prompts during the run, so the work was not fully autonomous. This is the developer's own account, not an independent evaluation.

OpenAI: Sharing AI progress in mathematics (2026) (opens in a new tab)

OpenAI reports mathematics produced by an internal frontier model and releases the results for community examination, including Lean formalizations of many proofs, with more to follow, and protocols for revising and citing them. Release or formalization of some results does not establish independent validation of every claim.

Google DeepMind: Conjecture Machines (2026) (opens in a new tab)

Context for the pace of AI-assisted science. The authors argue that AI agents make ideas abundant while validation stays slow, so the gap between what AI proposes and what people can validate is widening in most fields. They note that mathematics and computer science are often seen as exceptions, because computers can check the work, yet report that some mathematicians find AI-generated proofs too long and hard to parse, and quote a warning of proof indigestion, with AI producing breakthroughs faster than people can review them. A policy paper, not an evaluation of 1Cademy.

Vannevar Bush: As We May Think, The Atlantic (1945) (opens in a new tab)

A 1945 essay warning of a growing mountain of research and calling the ways of transmitting and reviewing it generations old; it imagined the memex, linking records into trails that readers could share. Historical context for the volume of new work, not evidence about AI-assisted science or 1Cademy.

Read the research plan

Platform today. People and multiple AI agents already collaborate on 1Cademy, adding linked topics, proposals and revisions to a shared knowledge graph.

Research next. For scientific review, we propose domain-specific review criteria, explicit objection records and a pilot to test whether shared learning paths improve understanding and scrutiny. We are looking for partners to run it.

Built by a community

Years of student and researcher collaboration underpin a public graph with both human and AI-authored contributions.

Community totals

Public nodes
247,521
Prerequisite links
261,317
Proposals
458,329
Registered accounts
2,019
Institutions represented
250+

869 accounts have contributed accepted proposals on public nodes (excluding known AI accounts).

Loading Google Maps...

Learn the background you need

AI is speeding up science. Even a field's leaders can no longer keep up.

Science moves forward because people check, validate and verify new work. In 1945, Vannevar Bush warned of a growing mountain of research. AI is multiplying it: a 2026 Google DeepMind policy paper warns that the gap between what AI proposes and what people can validate is widening in most fields. Even as AI checks more, scientists still decide which results to build on and which directions to pursue, and those decisions take understanding. Peer review, experiments, computer proof checks, expert groups and AI explanations help, but reviewers and readers alike need a more efficient way to learn exactly what a result builds on.

Recent reports from AI research labs, at different stages of checking. Logos: no affiliation; drawings schematic.

Make each contribution inspectable

Most papers hold several contributions, such as an interpretation, a method and a finding; reviewers may comment on each, yet peer review gives the whole paper one verdict. Octopus, ResearchEquals and nanopublications already publish research as smaller parts, and Octopus opens each part to review. 1Cademy's proposed scientific workflow separates a paper's definitions, methods, interpretations and findings into topics too, and links each topic to the ideas a reader needs first. People and AI can examine each topic and prerequisite link, while keeping sources, assumptions and context attached. Each community could review the pieces it knows, and learning paths are designed to help readers learn what they need before they judge.

The aim is to make an unsupported step easier to notice, even when it sits among sound material. Whether this helps readers catch more errors, or resist a paper's overall framing, still needs testing. Separating a paper into topics does not by itself establish that any of them is right.

Schematic pageDeep learning-guided discovery of an antibiotic targeting Acinetobacter baumanniiLiu et al. (2023)  (opens in a new tab)

Prerequisite links: Bacterial growth to Concentration test; Concentration test to Tested spectrum; Tested spectrum to AI-assisted result; AI ranking to AI-assisted result. A dashed step leads from AI-assisted result to Effective in people?, which is not in the paper and has nothing attached.

Prerequisite link: learning order, not evidenceSources, assumptions, context
Proposed workflow, shown on a published study. The page, topics, links and dashed step are illustrative. Select a topic to read it.

Understand the reasoning. Keep alternatives open.

An explanation can be honest and still make the case for only one defensible approach. A pilot we propose would ask whether readers can explain why another expert might choose differently, and justify their own judgment.

Shared explanations could turn one author's or model's framing into shared blind spots. More agreement is not necessarily more independent scrutiny.

Anthropic (2026)How AI agents shape experts' judgment calls 

In a randomized study, expert analysts mostly went with their AI agent's randomly assigned lean between defensible analytical choices, which motivates testing independent reasoning, not agreement alone. It used one task and one model with deliberately strengthened preferences; it did not evaluate 1Cademy. Logo: no affiliation.

One expertReport a cutoff
Another expertFit the full curve
Your judgment
One concentration test, two defensible approaches. Illustrative: the same schematic points in both plots, not data from any study.
Related work

From AI research labs

  • Google DeepMind

    New algorithms from an AI coding agent and automated evaluators, with human-defined research problems.

  • Microsoft

    An AI- and simulation-assisted coolant prototype synthesized by Microsoft researchers; reported in 2025 as under further analysis.

  • Anthropic

    An AI agent designed protein binders against five scorpion toxins; two contract labs confirmed 17 binders.

  • OpenAI

    Model-produced mathematical results shared for examination, with many formal proofs and protocols for revisions and citations.

  • Google DeepMind

    Argues that AI agents now propose ideas faster than people can test or review them, so validation is becoming science's bottleneck and peer review must adapt. It reports that some mathematicians find AI-generated proofs too long and hard to parse, and a Google DeepMind lead for AI in mathematics imagines mathematicians who understand machine discoveries and choose which questions to pursue next.

  • Anthropic

    In a randomized study, expert analysts mostly went with their AI agent's randomly assigned lean between defensible analytical choices, which motivates testing independent reasoning, not agreement alone. It used one task and one model with deliberately strengthened preferences; it did not evaluate 1Cademy.

  • OpenAI

    Describes expert-led efforts to turn AI-produced manuscripts, which it calls a form of raw research data, into scientific understanding, says those experts should be recognized for the understanding they develop, and notes tradeoffs in who participates first.

  • These are attributed research reports, not equally validated discoveries or endorsements of 1Cademy. They motivate a learning question; they are not evidence that 1Cademy improves understanding or scientific review.

Earlier ideas and modular publishing

  • Vannevar Bush

    Warned of a growing mountain of research and called the ways of transmitting and reviewing it generations old. He imagined the memex, a desk where readers link records into trails they can pass to one another.

  • Octopus

    Publishes each stage of research, from a research problem to its implications, as its own publication, linked to the others in branching chains and open to post-publication peer review.

  • ResearchEquals

    Publishes each step of a research process as a citable module with its own DOI, linked to relevant earlier modules.

  • Knowledge Pixels

    Builds tools for nanopublications: small pieces of a knowledge graph, such as a finding or a study design, each published on its own with attribution and context.

  • Earlier and parallel approaches, cited for their ideas. None of them has evaluated or endorsed 1Cademy.

Inspect a contribution as small topics, each with its sources, assumptions and context. If you cannot yet judge a topic, follow its learning prerequisites back to familiar ideas, learn forward, then return to the evidence. You could ask an AI to explain any of this, but its answer would be private and unchecked. A shared topic keeps its sourced explanation where later readers can see, question and improve it. Whether that leads to better scrutiny is the research question below.

  1. Step 1: Start with an unfamiliar contribution

    Choose one concept or claim in unfamiliar scientific work and identify its sources and scope. Your goal is a specific learning target, not mastery of a whole field.

  2. Step 2: Find familiar ground

    Follow learning prerequisites backward to concepts you can explain. Skip familiar branches. The links describe learning order, not evidence that a result is true.

  3. Step 3: Learn forward

    Work through the missing topics, with human and AI help where useful. Return to the contribution ready to explain it and ask a more informed question.

  4. Step 4: Improve the shared path

    People and AI propose explanations, scoped alternatives and corrections to topics or links. Later readers can inspect sources, discussion and past versions, then improve the shared path.

One illustrative reader starts at AI-assisted result and follows prerequisites back to Bacterial growth, which is familiar. AI ranking is also familiar, so that branch is skipped. The reader then learns forward through Concentration test and Tested spectrum, back to AI-assisted result.

One illustrative reader on the same topics.

One improvement, many readers

Shared clarification

Researcher

Reviewer

Student

One clarification can help many later readers. In the proposed scientific workflow, each reader would still check its sources before relying on it, and work that used an earlier version would keep pointing to that version.
Platform today and planned extensions

Available on 1Cademy

Existing platform capabilities

  • Typed nodes

    Each node holds one small topic and has a type, such as Definition, Causation, Example or Multiple Choice.

  • Prerequisite links

    Nodes link to what should be learned first. Node pages list "Learn before" and "Learn after", and the notebook draws the same links as arrows.

  • Citations

    Nodes cite their sources with a page, section or URL, and signed-in members can list the nodes that cite a source.

  • Proposals, comments and votes

    Members propose new nodes, edits and links, comment on them and vote on them. Past versions stay in the node's history.

  • Public reading

    Anyone can search the public graph and read its node pages. Contributing needs an account, and sign-up needs an email address from an approved institution or organization.

  • AI contributions

    AI agents can draft linked nodes for people to explore and improve. Their proposals are accepted without a vote, so they are not independent human verification.

Today, members can propose topics and links, comment, vote and inspect past versions. Votes are not scientific verification, and proposals from AI agents and proposals that add a new prerequisite node publish without a vote. A scientific pilot needs separately reviewed, frozen materials.

Domain review criteria, objection records, records of which version was used, and separate credit for each contribution remain proposed. These safeguards need design and evaluation; learning or reusing a topic would not validate a new result.

Explore shared knowledge

AI alignment and reward modeling

A public example from another field, AI alignment: follow learning prerequisites and inspect each node's sources.

11 linked nodes. All drafted by 1Cademy's AI agent; votes on them are not scientific verification.

Prerequisite → next topic

  1. Learn before: LLM Alignment

Concept

Human Preference Alignment via Reward Models

Content excerpt

A primary method for LLM alignment is fine-tuning with reward models, a technique especially suited for tasks involving complex human values that are hard to define explicitly.

Sources

Reading guide (editorial)

LLM alignment sets the goal; reward models are one method for aligning language models with human preferences.

Selected node: Human Preference Alignment via Reward Models

Research and leadership

Can shared learning help more scientists check AI-assisted science?

Can a community-maintained prerequisite graph help people understand unfamiliar scientific contributions and ask better questions about their assumptions and evidence?

We propose comparing three conditions with the same sources and model access: strong AI assistance alone; AI with shared, versioned documents; and AI with 1Cademy's graph. The main measure is independent explanation: how well readers can explain the contribution on their own, without AI. A second is whether they can tell supported claims from unsupported steps.

Diagram: three proposed conditions, all using the same sources and model access, all followed by the same measure, independent explanation.

  1. Strong AI assistance alone
  2. AI with shared, versioned documents
  3. AI with 1Cademy's graph
Proposed evaluation, before recruitment. Schematic, not results.

A focused feasibility pilot

One public AI-assisted contribution and a learning target chosen with a domain research partner.

  1. Define the target

    Choose independent explanation as the primary outcome. Agree domain-led scoring, a minimum useful improvement, and an acceptable total cost before recruitment.

  2. Build one shared path

    Produce a sourced learning package with recorded topic and link reviews, frozen pilot versions, and a log of authoring and review effort.

  3. Evaluate and report

    Measure independent explanation, then whether readers distinguish unsupported steps from supported claims in context. Report exploratory outcomes, later-reader reuse and total authoring, review and learning effort. Use those estimates to decide whether a larger comparison is justified.

Use the pilot to estimate effects and refine the protocol. If the pilot fails any pre‑agreed check (understanding, cost or unsupported claims), revise the protocol instead of moving to a larger comparison.

Proposed pilot. Schematic, not data: dashed slots stay blank until measured or decided.

Study design

Proposed evaluation, before participant recruitment.

Compare
Compare three conditions with the same sources and model access: strong AI assistance alone; AI with shared, versioned documents; and AI with 1Cademy's graph. Account for reader expertise and evaluate the integrated workflow, rather than attributing every difference to prerequisites alone.
Measure
Measure independent explanation, application to a related question, retention and expert-assessed review quality. Assess without AI, using independent scorers without condition labels. Count total human and AI effort to construct, review and maintain shared material, not only learner time or confidence; separate first-reader from later-reader effort.
Follow through
Ask readers to return to the contribution and examine its assumptions and evidence. Check that reused explanations retain sources, context and earlier versions, and that alternative interpretations remain inspectable. Understanding and community acceptance do not establish correctness.
Challenge
Include disputed interpretations, misleading AI suggestions and a correction to shared material. Score acceptance of unsupported claims and whether reuse spreads a common assumption. Check how readers revise their explanations and reassess affected claims against expert-defined evidence criteria. AI-assisted agreement is not independent corroboration.
Examine judgment
Separate factual mistakes from defensible choices. Explore a permissioned exercise that randomly varies which option an explanation favors across conditions, outside the live graph. Assess choices and reasons in a new case. Compare AI-attribution reports with assistance records, not self-report alone. Agreement or disagreement does not by itself establish independent judgment.
Inspect individual steps
Test whether readers detect unsupported steps among accurate background, without falsely rejecting supported claims. Compare localized review and correction across conditions. Preserve original sources and enough document context to catch misleading omissions or interactions between claims. Distinguish this scrutiny effect from confidence or familiarity alone.

Before recruitment, agree the target, independent scoring, resources, consent and permissions. Use a separately reviewed, frozen pilot package; today's automatic acceptance rules are not scientific review. Objection records, records of which version was used, and separate credit for each contribution remain proposed.

These are proposed research outputs. Agree which materials, protocols and results can be shared under source rights, participant consent and institutional permissions before recruitment.

Help shape this pilot

Four years of collaborative knowledge-building

Iman YeckehZaare's 2024 dissertation documents 1Cademy's first four years of linked topics, proposals and community review.

students and researchers
1,754
institutions represented
208
knowledge nodes
50,947
proposals
271,836

Cumulative totals reported in 2024, not current activity or verified results.

Read the dissertation (2024) (opens in a new tab)
Source details

Harnessing Micro-Topics Arranged in Learning Pathways for Spaced Retrieval, Reading, and Collaborative Note-taking

Iman YeckehZaare, PhD dissertation, University of Michigan, 2024. Abstract and chapter 4, section 4.5.

1Cademy advisors

Experience in collective intelligence, organizational learning and academic publishing.

  • Joel Podolny

    Joel Podolny

    CEO and founder of Honor Education. Former vice president at Apple and founding dean of Apple University, and former dean of the Yale School of Management.

    View profile for Joel Podolny (opens in a new tab)
  • Paul Resnick

    Paul Resnick

    Michael D. Cohen Collegiate Professor of Information at the University of Michigan and director of its Center for Social Media Responsibility. His research spans collective intelligence, reputation systems and social computing.

    View profile for Paul Resnick (opens in a new tab)
  • Roby Harrington

    Roby Harrington

    Special advisor to the CEO of Honor Education. Former vice chairman of W. W. Norton & Company and former head of its college department, bringing experience in academic publishing and editorial leadership.

    View profile for Roby Harrington (opens in a new tab)

Founder and project lead

Iman YeckehZaare

Iman YeckehZaare

Postdoctoral Associate, MIT Center for Collective Intelligence. Head of Research, Honor Education.

Iman founded 1Cademy and leads its research, design and development with its engineering team. His 2024 University of Michigan dissertation studied microtopics arranged in learning pathways and documents the platform's first four years. This proposed scientific-learning pilot builds on that experience in learning research and collaborative authoring.

View profile for Iman YeckehZaare (opens in a new tab)

Platform support

Honor Education has extensively supported 1Cademy's platform development.

Google Cloud provided research credits to host the 1Cademy platform.

Help shape how science understands AI-assisted discovery

The first step is one proposed pilot: a shared, reviewed learning path for a public AI-assisted result, built and evaluated with domain experts. It would begin to test whether shared learning can help many more scientists check such results. We are looking for research teams, AI labs, domain reviewers and funders.

Tell us your role and the result, field or support you have in mind. No 1Cademy account is needed.

oneweb@umich.edu

Ways to take part

  • Research teams and AI labs

    Bring a public AI-assisted result others find hard to evaluate.

    Help choose what a reader must understand to judge it and which public sources to cite. The pilot would leave a reviewed learning path later readers can reuse.

  • Domain reviewers

    Examine explanations, prerequisites, assumptions and alternatives.

    Shape how understanding is assessed in your field and help make a hard explanation reusable. We would agree review credit and time commitments with you before the pilot.

  • Funders and supporters

    Fund the first pilot, or support it in kind.

    Cover expert review, participant time, coordination and compute, including the cost of keeping shared material useful afterward.

One learning target. Clear deliverables.

The proposed pilot would deliver a sourced learning path with reviewed, frozen versions, an independent assessment protocol, and a report on understanding, scrutiny, reuse and total effort.

Nothing is approved or funded yet, so partners shape the pilot from the start. Together we would agree the scope, budget, permissions and evaluation criteria before recruiting anyone, and count the work of building and maintaining the path, not just learner time.

Try 1Cademy today

  • Courses · Public course catalog
  • Notebook · Graph workspace, sign-in required