Stop, Collaborate, and. . . Lessons Learned from Collaborative Human-AI System Development
Bibliographic record
Abstract
The U.S. military is procuring systems for complex joint fighting, requiring novel teaming approaches that maximize the capabilities of humans, artificial intelligence (AI), and autonomy. This case study describes a research effort to support aerial reconnaissance and target acquisition with crewed and uncrewed platforms. This effort integrated human factors with cutting-edge AI techniques, using Decision-Centered Design to identify requirements of the reconnaissance work and multiple AI development approaches. Lessons learned include the following. First, the team applied Roth and colleagues’ macrocognitive synthesized framework of decision making. Its emphasis on integrated human and AI sensemaking capabilities, as well as interfaces to facilitate common ground, facilitated cross-discipline collaboration. Second, the team collaborated on the goals and data of underlying AI features in parallel with human-AI interaction development. Third, scenario-based design was valuable to facilitate co-design activities. Embedding cognitive requirements within scenario storyboards made complex military work accessible to the full engineering team.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.030 | 0.032 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.004 | 0.012 |
| Scholarly communication | 0.009 | 0.014 |
| Open science | 0.003 | 0.006 |
| Research integrity | 0.003 | 0.005 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".