SOSS and Loom: generative AI as an active creative medium

A creative-process framework, SOSS (Shape, Observe, Stir and Select), treats generative AI as an active medium rather than only an advisor. The authors introduce Loom, a creative-writing probe in which users orchestrate simulated narrative agents.

Workflow level n/a Source Mixed sources Outcome Not empirically evaluated here

Evidence & verification

Source level

Mixed sources
Grounded in a combination of peer-reviewed and non-peer-reviewed sources; see the individual sources.

Outcome verification

Not empirically evaluated here
A documented resource or intended workflow, not a demonstrated academic outcome. No empirical benefit is established by its inclusion.

Sources checked

2026-10-17 (UTC)

Primary scope: Material for Thought — primary arXiv v1 abstract: abstract-only; CHI 2026 Tools for Thought — official accepted-submission list: partial-documentation. Publication classification is separate from substantive read scope; see limitations and entry-audit.json.

Source-checked does not mean independently reproduced. See the Method page for how these labels are assigned.

Sources (2)

Research applicability

A concrete interaction-design reference for arts and HCI research into AI-assisted creative writing and human orchestration of narrative agents. Creative writing is not relabelled scientific-writing.

Getting started

Editorial, untested: read the SOSS description and use its four modes to discuss a supervised creative-writing session. Consult the full Loom design before treating this as an executable interaction recipe.

Maintenance

Sources accessed 2026-10-17; service availability, maintenance and compatibility were not tested.

License

Reuse permissions for paper, data and code require separate assessment; no blanket licence or redistribution assurance.

Workflow & human oversight

Resource-level workflow level intentionally unspecified. Conceptual references and reusable process patterns do not themselves establish an autonomous workflow; researchers retain task-specific interpretation and oversight.

Workflow level is not an evidence score; higher is not better evidence.

Approach

Drawing on Schön's reflective practice, the authors frame ongoing human–AI conversation through SOSS and argue that human disruption and curation matter when generative systems tend toward convergence. This description is conceptual, not a verified creative-quality result.

Limitations

Abstract-only substantive review with official workshop acceptance-list metadata. The full Loom design, implementation, access terms, prompts, experiments and usability were not read or tested. A CHI 2026 Tools for Thought non-archival workshop contribution is confirmed; the site's acceptance list does not itself establish peer-review status. Mixed denotes arXiv abstract and official workshop documentation, not confirmed archival peer review. Claimed creative quality and convergence tendencies are the authors' arguments, not reproduced outcomes.

Reproducibility

AI-agent primary-source reading and provenance checks only. No software, experiment, hardware, outcome or installation reproduced. Source verification is not outcome verification.

Adoption requirements

Editorial, untested: obtain the complete description and establish whether the probe is accessible before attempting implementation. Use material with appropriate rights and protect private writing; neither a software distribution nor a working installation was verified.

Comments

No comments yet.

Add a comment

Comments are held for moderation before appearing. Critical or corrective comments are welcome — especially evidence that an entry overstates what has been verified.