Productivity loss
End user's loss of productivity due to the underperfomance of a genAI application, including producing nonsensical or poor quality outputs, degrading its utility.
ENTITY
2 - AI
INTENT
2 - Unintentional
TIMING
2 - Post-deployment
Risk ID
mit1349
Domain lineage
7. AI System Safety, Failures, & Limitations
7.3 > Lack of capability or robustness
Mitigation strategy
1. Implement robust, multi-layered quality control mechanisms, including post-generation "hard guardrails" to score factual accuracy and compliance, and standardize prompt management (e.g., version control, grounding instructions) to minimize nonsensical or poor-quality outputs and system failures. 2. Develop and enforce a comprehensive, differentiated training strategy for end-users that emphasizes "verification literacy," critical evaluation of AI outputs, and clear operational guidelines for when and how to appropriately apply the AI tool to mitigate automation bias. 3. Conduct a strategic zero-based redesign of end-to-end workflows, instrumenting processes with dual measurement—pairing usage analytics with objective quality-of-output metrics (e.g., error rates)—to accurately track performance, convert potential gains into captured business value, and inform targeted AI deployment.