> ## Documentation Index
> Fetch the complete documentation index at: https://irisdocs.prescientlabs.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# AI Risk & Governance

> Identify material AI risks early and build proportionate controls into recommendations without turning every engagement into a legal project.

AI risk is part of good consulting. It should influence whether an opportunity is recommended, how it is designed and what human oversight is required.

## Review risk across six dimensions

1. **Data:** personal, confidential or sensitive information.
2. **Accuracy:** consequence of incorrect output.
3. **Fairness:** potential impact on people or protected groups.
4. **Security:** access, data leakage, prompt injection and tool permissions.
5. **Operational control:** human review, fallback and monitoring.
6. **Regulatory context:** sector, geography and use-case obligations.

## Risk should affect priority

A technically feasible opportunity may still be a poor first project if:

* an error could materially harm a customer or employee;
* required data cannot be handled appropriately;
* there is no owner for ongoing monitoring;
* the process is highly regulated;
* the proposed autonomy exceeds the organisation’s control maturity.

## Keep governance proportionate

A low-risk internal drafting assistant does not need the same governance as an automated employment or financial decision system.

## Consultant responsibility

You should surface risk and involve qualified legal, security, privacy or regulatory specialists where needed. This hub is practical education, not jurisdiction-specific legal advice.

## What to do next

Read **Responsible AI**, **Data & Privacy** and **Client Governance**.