New white paper - AI Guardrails: concepts, models and methods

News | Posted on Friday 20 December 2024

In our latest white paper we look at a growing trend in discussions around risks in AI models - that of guardrails.

Yellow arrow road signs show a steep bend in the road with safety barrier
Image by Veronica Bosley from Pixabay

The concept of "guardrails" as a means of controlling the risks associated with AI is a common one. However, is this an over-simplistic view. How can we draw on insights from the safety engineering sector to create more specific and detailed risk control strategies for AI - especially General Purpose AI (GPAI)?

Download our AI Guardrails White Paper (PDF , 1,270kb) to learn more.