A vendor policy guaranteeing your inputs and outputs are not stored, logged, or used to train future models beyond the immediate request. The non-negotiable baseline for research with participant data.
Definition: A vendor policy guaranteeing your inputs and outputs are not stored, logged, or used to train future models beyond the immediate request. The non-negotiable baseline for research with participant data.
A vendor policy guaranteeing that your inputs and outputs are not stored, logged, or used to train future models beyond the immediate request. The non-negotiable baseline for any research touching participant data. If a vendor cannot point to a written ZDR clause, assume your data is training material. Covered in detail in Evaluating AI Research Tools.
An attack class where untrusted input carries instructions that hijack a model's behavior. Any AI system ingesting external content (transcripts, tickets, scraped pages) is a security surface.
AI-generated personas marketed as substitutes for real research participants. The pitch is faster and cheaper than recruitment; the reality is regression to training-data averages dressed up as user voices.
A SaaS product built on top of a foundational model, adding a UI, workflow, and usually a hidden system prompt. You get speed; you give up control.