Topic 10 of 20 · Part III : Personalisation and Responsible AI

Safety and guardrails

Guardrails should be layered: a prompt alone is too fragile, while a rigid rule set alone cannot interpret every context.

The control pattern

Data asset

Risk examples, policies and runtime checks

Knobs

Risk categories, Coverage, Thresholds, Adversarial cases, Permissions, Escalation rules

Outcome

Safer outputs and actions

Measurement

See how to measure this below

What it is

Safety guardrails define what the system should refuse, qualify, verify or escalate. The data asset includes risk-labelled examples, explicit policies, deterministic checks and runtime signals.

Knobs include which risk categories are covered, how conservative the thresholds are, what actions require confirmation, and how adversarial or manipulative inputs are handled. Guardrails should be layered: a prompt alone is too fragile, while a rigid rule set alone cannot interpret every context. Combining policy-aware models with structured validation and permissions produces a more dependable boundary.

Overly permissive systems create harm; overly restrictive systems drive users to ignore warnings. The best guardrail explains the boundary in plain language and redirects toward verifiable information.

The knobs in detail

Each row is one adjustable property of the data asset, and what moving it tends to do.

KnobWhat you adjustLikely effect
Risk categoriesWhich harms are explicitly coveredDefines the boundary being enforced
CoverageHow much of the surface is checkedCloses gaps between policy and product
ThresholdsHow conservative each check isBalances harm against false refusals
Adversarial casesManipulative and injected inputsHardens the system against source text
PermissionsWhat requires explicit confirmationKeeps consequential actions deliberate
Escalation rulesWhen a human or larger model steps inProvides a safe path instead of a refusal

Applied: Stocks Assistant

For Stocks Assistant, safety controls should prevent fabricated certainty, unqualified guarantees, unsupported price targets, and claims that a signal is personalised advice when suitability data is missing. Retrieved documents can contain malicious or irrelevant instructions, so the system should treat source text as evidence rather than executable commands. A research assistant should be able to explain risk, but direct trade execution or account changes should stay outside the workflow unless separately authorised with clear confirmation and auditability. High-volatility products and options deserve stronger warnings than general educational content.

How to measure it

Evidence that the knob produced the intended behaviour, rather than shifting the problem elsewhere.

  • Policy-violation rate on ordinary and adversarial tests
  • False refusal rate
  • Escalation quality : is the safe alternative still useful?
  • Auditability of any consequential action

Common mistakes

Relying on prompt instructions alone as the safety layer.

Treating retrieved document text as instructions rather than evidence.