human escalation
A mechanism in AI-assisted systems that automatically transfers requests exceeding preset boundaries to human handling. It typically combines confidence thresholds, keyword rules and issue classification models, and in high-risk domains such as legal must be designed conservatively enough.
Timeline (last 90 days)
In AI-assisted legal systems, human escalation is a safety mechanism: when the system identifies that a question exceeds preset boundaries, it automatically forwards the request to a human lawyer.
Opening legal knowledge to business departments carries the risk of misinterpretation, so redesigning the intake stage must include boundary design that specifies which questions the system can answer directly and which must be escalated to a human.
All Facts (2)
In AI-assisted legal systems, human escalation is a safety mechanism: when the system identifies that a question exceeds preset boundaries, it automatically forwards the request to a human lawyer.
50%UnverifiedOpening legal knowledge to business departments carries the risk of misinterpretation, so redesigning the intake stage must include boundary design that specifies which questions the system can answer directly and which must be escalated to a human.
50%