For research & standards collaborators
The Human Intelligence Layer raises questions worth examining together. What makes human context useful? What motivates people to contribute? And what must travel with a record for someone else to understand it?
We welcome researchers, standards contributors, learning institutions, evaluators, and ecosystem practitioners who can help turn those questions into focused investigations and practical tests.
What benefits encourage useful participation? How do people understand, correct, and control representations of their experience?
Does connecting workflow context, experience, and evidence help people identify gaps or coordinate decisions? Under what conditions?
What context must accompany a learning or employment record so a receiving party can interpret it appropriately?
How should source evidence, uncertainty, interpretation, and human review remain visible when AI helps organize information?
Challenge definitions, assumptions, and causal claims. Help identify where the argument needs stronger evidence.
Define a research question, appropriate method, measures, and limitations around a bounded use case.
Examine a specified record exchange, implementation behavior, or interpretation problem with participating systems.
Explore jointly authored guidance, mappings, examples, or findings with attribution and reuse agreed in advance.
Imagine two systems successfully exchanging a learning record. The receiving organization can read it, but still cannot tell whether the evidence is relevant to its decision.
A joint investigation could examine what additional context is needed, how people interpret it, and whether a changed presentation helps. That separates transport and verification questions from the practical question of usefulness.
This is a proposed research direction, not a reported result. Success criteria, methods, and limits would be defined before testing.
Specify the question, participants, methods, measures, access requirements, and any necessary ethics review.
Distinguish observations from inferences. Record limitations and findings that challenge the initial hypothesis.
Agree on authorship, attribution, confidentiality, review, funding, and rights to publish or reuse materials before work begins.
We want findings to improve the model, the product, and the way people work with it. Partners can help us identify which problems require better configuration, stronger evidence, or a change in the underlying design.
A collaboration does not imply standards-body endorsement, certification, or product compliance. Any compatibility claim must be tied to the specific implementation and test performed.
No. Useful research can qualify, challenge, or reject an assumption. The research question and reporting arrangements should make that possible.
Access requires a defined purpose, appropriate authorization, and agreed protections. Participation in a collaboration grants no automatic access.
We can explore a defined contribution. Authorship, participation rules, publication, and any submission authority need to be agreed.
Resourcing is considered for each project. We can discuss a bounded contribution or a funding proposal, but no standing grant program is promised.
Start with these resources and bring the questions they raise to our conversation.
Review the definitions and design direction we want to examine.
Distinguish proposed architecture from available capabilities.
We’re building this with early collaborators. Available capabilities, configuration, integrations, and support are confirmed for each engagement. The configurations described here include development directions.
Bring a research question, interoperability challenge, or proposed contribution. Tell us the setting, evidence, and collaborators that could make it a useful investigation.