Engineering a Governance-Aware AI Sandbox: Design, Implementation, and Lessons Learned
arXiv 2603.03394•a270dc74934d8a55ce7fc3fd4d69494083f9ea0396aeb59353d46b99bbd3738b
AI governanceLLM code generationLoRAPEFTadversarial evaluationaudit loggingbenchmarkingcode synthesisconcurrencycontinuous integrationdeadlockensemble methodsforensicsgoal driftmaintainabilitymodel attributionmodel compositionmulti-tenantprivacyprompt/agent alignmentrace conditionreverse engineeringreviewer patternsandboxsoftware protection
Paper metadata
- arXiv ID
- 2603.03394
- Version
- Not specified by this published record
- Category
- Computer Science — Software Engineering (cs.SE)
The PDF link points to arxiv.org. Baitaphish does not expose a private stored PDF.
Evidence and limitations
- Source ID
- arxiv_cs_se
- Record identifier
- a270dc74934d8a55ce7fc3fd4d69494083f9ea0396aeb59353d46b99bbd3738b
- Enrichment time
- 2026-03-05T13:37:54Z
- AI-assisted enrichment
- Yes
This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.