From the hash-bound evidence releases
Recorded claims
Each claim states what the inspected source says, at the recorded location—bounded by its scope and graded by its confidence. Nothing here is a synthesis across works.
The study reports that more than 10% of observed users ran three or more agents concurrently and 26.6% used reusable skills, documenting emerging delegation and reusable-workflow behavior rather than merely chat-style assistance.
Scope: Observed Codex users during the study period; behavioral adoption, not output correctness or economic value.
Agent use became nearly universal inside OpenAI but remained lower and uneven across external organizations; the paper's output-token and usage measures cannot establish productivity, quality, or optimal organization structure.
Scope: OpenAI internal use versus external Codex customers; selected organizations and metrics.
Structured relationships
In-corpus citations
Only source-supplied cites relationships whose two endpoints are admitted are shown. Invocation evidence remains a separate relationship.
Cites (0)
No in-corpus outgoing citation is available in the current enrichment coverage.
Cited by (0)
No in-corpus incoming citation is available in the current enrichment coverage.
Boundaries
Limitations & independence
Recorded at coding time, carried with the work forever. A claim without its limits is not evidence.
Recorded limitations
- Vendor-controlled telemetry
- Output tokens and session patterns are not productivity or quality
- Coding-agent users and organizations are selected
- Fast-changing product capabilities
- Preprint
Source independence
OpenAI-affiliated analysis of Codex telemetry, including OpenAI as an internal case; not independent of product or focal organization.