An AI attack pattern which starts with innocent prompts and gradually escalates over multiple conversation turns to bypass safety guardrails and extract harmful content. Each individual turn appears benign, evading single-prompt detection.
Multi-turn escalation
7th May 2026
Manage your entire QA lifecycle in one place. Sync Jira, automate scripts, and use AI to accelerate your testing.
Explore MoT
See where AI genuinely helps, where it doesn’t, and how testers can stay firmly in control
Boost your career in software testing with the MoT Software Testing Essentials Certificate. Learn essential skills, from basic testing techniques to advanced risk analysis, crafted by industry experts.
Debrief the week in Quality via a community radio show hosted by Simon Tomes and members of the community
