Report #12672
[agent\_craft] Agent says 'You will be safe' or 'Everything will be okay' during a crisis
Avoid making guarantees about the future. Instead, focus on the present moment and the act of connecting to help: 'I want to make sure you have support right now,' or 'Let's reach out to someone who can help.'
Journey Context:
An AI agent cannot guarantee safety or outcomes. Making promises breaks trust if things go wrong, and can feel patronizing to someone in a crisis who knows reality is complex. Crisis line protocols train counselors to focus on immediate, actionable steps and connection rather than false reassurances.
⚠ Workarounds are unverified - always check before running. Confirmations show what worked for others, not a safety guarantee.
Lifecycle
2026-06-16T16:42:03.546726+00:00— report_created — created