Policy as enforced rules: approved models by task, data type and risk.
Rules turn policy into enforced rules, including the approved model by task, data type and risk, and where AI must not be used. Guard enforces them, so premium is by exception. Plain-English authoring, conflict checks and suggested rules are being built.
Most AI policies are documents. Documents do not stop a request.
The policy is followed when convenient and breached quietly, and you cannot show which.
of employers have an AI policy at all
CIPD, 2026The benefits of Rules.
The people who feel the problem first.
Turn policy into something that runs.
Know the rules the business set are the rules that apply.
Maintain rules in one place.
Three ways teams use Rules.
Block uploads of data that must not leave, and notify the owner.
Send routine work to standard models by rule.
Keep AI out of the decisions it must not make.
Works best alongside.
A tracked catalogue of models, tested before they reach your people.
See Change Control Data ProtectionSensitive data held back before a request reaches a model.
See Data Protection LogLog: the record of events.
See Log All of GuardSystems, controls and process designed with AI agents and aligned to the regulations, your policies and outside policies. Used with Evidence.
See the Guard modulePrices are indicative ranges, confirmed in writing before you commit. Estimates are replaced by your own figures once your health check runs.
How the supporting tools fit together
GenWA Frame brings together your policies, business knowledge and our methodology to guide the work.
The Log records what connected AI saw, produced and who approved it.
The Intelligence Layer is being built to turn those records into trends and answers linked to their source. We agree which tools are available and appropriate for your engagement.
Find out more about the GenWA FrameSee Rules working on your business.
Start with a health check. Agree the scope, fees and next steps with us.