Direction, evidence grade, and study type are AI-generated labels (gpt-5-mini), not human-verified. Syntheses are LLM-written. "Tensions" are machine-detected candidates, not confirmed contradictions. A research-acceleration tool, not peer review.
How this is built →
2Distinct papers
4Unique collaborators
2/2Semantic Scholar citation coverage
Publication span: 2026. Corpus fetch span: 2026.
Identity provenance
Provider IDs
- Semantic Scholar:
2065557640
ORCID evidence
No valid ORCID is stored.
Observed aliases (1)
- Emad Shihab (semantic scholar, provider refresh)
Topics and outcomes in this view
Assessment themes
- Human Ai Collab: 2 papers
- Productivity: 2 papers
- Org Design: 1 paper
Claim outcomes
- Organizational Efficiency: 2 papers
- Adoption Rate: 1 paper
- Error Rate: 1 paper
- Governance And Regulation: 1 paper
Papers in the Semantic Scholar view
Latest stored Semantic Scholar author observations only. Citation counts below are from the same provider and are not combined with other services.
Scroll the table horizontally to see every column.
| Paper | Author evidence | Date | Provider citations |
|---|---|---|---|
| AI coding assistants produce code that tends to stick: agent-authored lines are modified less often and face a 16% lower hazard of change than human-written lines, but they show modestly elevated corrective edits and greater variation across agents — implying organizational practices, not generation quality alone, shape long-term maintenance.arxiv | Emad Shihab provider id |
2026-01-23 | 3 |
| Agent-based software engineering spends most tokens refining code, not generating it: in 30 ChatDev GPT‑5 projects the Code Review stage consumed 59.4% of tokens while input tokens made up 53.9% of usage, indicating iterative agent collaboration is the primary operational cost driver.arxiv | Emad Shihab provider id |
2026-01-20 | 8 |
Citation observation summary
Semantic Scholar supplied counts for 2 of 2 papers in this view; 0 are missing. The observed paper counts sum to 11 cumulative citations. This is a coverage summary, not an author score or h-index.