Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Refreshing forked repository with recent changes from Rob #51

Open
wants to merge 37 commits into
base: main
Choose a base branch
from
Open
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
37 commits
Select commit Hold shift + click to select a range
e2dcf4f
Update owaspaiexchange.md
rkopaee Feb 12, 2024
6d8b24f
Update owaspaiexchange.md
rkopaee Feb 12, 2024
18903d4
Update owaspaiexchange.md
rkopaee Feb 12, 2024
8f5f02d
Update owaspaiexchange.md
rkopaee Feb 12, 2024
1989063
Update owaspaiexchange.md
rkopaee Feb 12, 2024
c1e56d9
testing image upload
rkopaee Feb 12, 2024
4e4a5e8
Add files via upload
rkopaee Feb 12, 2024
2eec559
Update owaspaiexchange.md
rkopaee Feb 12, 2024
392677c
Update owaspaiexchange.md
rkopaee Feb 12, 2024
62a857b
Update owaspaiexchange.md
rkopaee Feb 12, 2024
6b5efb4
Update owaspaiexchange.md
rkopaee Feb 12, 2024
75013a3
Update owaspaiexchange.md
rkopaee Feb 12, 2024
aeecdd6
Update owaspaiexchange.md
behyka Feb 12, 2024
42ff456
Merge pull request #2 from behyka/patch-4
rkopaee Feb 12, 2024
3b2e0a3
Merge pull request #1 from OWASP/main
rkopaee Feb 16, 2024
28243b5
Riskmanagement1
behyka Feb 16, 2024
49c473c
Update owaspaiexchange.md
behyka Feb 16, 2024
c8f9398
Update owaspaiexchange.md
behyka Feb 16, 2024
ede80fa
Update owaspaiexchange.md
behyka Feb 16, 2024
4e91c2c
Update owaspaiexchange.md
behyka Feb 16, 2024
e73477a
Update owaspaiexchange.md
behyka Feb 16, 2024
1ae526f
Update owaspaiexchange.md
behyka Feb 16, 2024
c91b066
Update owaspaiexchange.md
behyka Feb 16, 2024
1ec6db3
Update owaspaiexchange.md
behyka Feb 16, 2024
489b0d6
Riskmanagement2
behyka Feb 16, 2024
52844e8
Update owaspaiexchange.md
behyka Feb 16, 2024
1b6eacb
Update owaspaiexchange.md
behyka Feb 16, 2024
1214b93
Riskmanagement3
behyka Feb 16, 2024
f2b5d71
Update owaspaiexchange.md
behyka Feb 16, 2024
189d255
Update owaspaiexchange.md
behyka Feb 16, 2024
360d588
Update owaspaiexchange.md
behyka Feb 16, 2024
7ae74bd
Update owaspaiexchange.md
behyka Feb 16, 2024
9847118
Update owaspaiexchange.md
behyka Feb 16, 2024
e25331f
Update owaspaiexchange.md
behyka Feb 16, 2024
3717ba1
Update owaspaiexchange.md
behyka Feb 16, 2024
cfd5804
Update owaspaiexchange.md
behyka Feb 16, 2024
42080b8
Update owaspaiexchange.md
behyka Feb 16, 2024
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Binary file added assets/images/RiskManagement2.png
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added assets/images/Riskmanagement1.JPG
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added assets/images/Riskmanagement2.JPG
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added assets/images/Riskmanagement3.JPG
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added assets/images/image.png
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
91 changes: 91 additions & 0 deletions owaspaiexchange.md
Original file line number Diff line number Diff line change
Expand Up @@ -49,6 +49,7 @@ Table of contents:
- [2. Threats through use](#2-threats-through-use)
- [3. Development-time threats](#3-development-time-threats)
- [4. Runtime Application security threats](#4-runtime-application-security-threats)
- [5. Risk Management](#5-risk-management)
- [References](#references)
- [Expanded Table of contents](#expanded-table-of-contents)

Expand Down Expand Up @@ -1439,6 +1440,94 @@ GenAI models mostly live in the cloud - often managed by an external party, whic

- **#MODELINPUTCONFIDENTIALITY** (runtime appsec). Model input confidentiality: see SECDEVPROGRAM to attain application security, with the focus on protecting the transport and storage of model parameters (e.g. access control, encryption, minimize retention)

# 5. Risk Management

---

## 5.1. Introduction
AI Risk Management applies regardless of where in the supply chain your organization fits.
Whether you are integrating AI systems built by others or developing ,modifying or training them to be utilized internally or externally by others, it is imperative to manage their risks.

In the context of AI Risk Management process, the term AI encompasses all flavors of AI including AGI, Machine Learning, Deep Learning, and Data Science, while acknowledging that the actual risks overlap and vary depending on the flavor of AI and the applicable domain.

The intent of this document is to concisely capture the key elements of an effective AI risk management framework and provide a prescriptive framework. This document is organized by AI Risk Management Process, Model, Activities, Risk Catalog, Roles and Responsibilities and Risk Catalog.

## 5.2. AI Risk Management Process

We utilize the ISO 31000 as a good framework to overlay our AI Risk Management process.


<img src="assets/images/Riskmanagement3.JPG" align="left" width="800" height="500">


<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
<br>
Source: ISO 31000:2009(en), Risk management — Principles and guidelines








## 5.2.1. Establishing Context
<img src="assets/images/Riskmanagement1.JPG" align="right" width="250" height="400">

- **External Context**: Understand the environment in which ML, Data Science, and Gen AI systems operate, including legal, regulatory, financial, and social factors. Consider compliance with privacy (e.g. GDPR in Europe), or specific jurisdictional (e.g. Canada Bill C27-AIDA) or industry (e.g. Healthcare or Finance) requirements.
- **Internal Context**: Assess the capabilities, culture, and objectives of the organization developing or using these technologies. This includes technical infrastructure, data governance policies, ethical standards, and the strategic goals related to AI deployment.
- **Risk Management Context**: Define the scope and objectives of risk management activities, considering the stakes involved in deploying AI systems, such as ethical implications, bias, and potential for misuse.
<br>
<br>
<br>
<br>



## 5.2.2 Risk Assessment
<img src="assets/images/Riskmanagement2.JPG" align="right" width="250" height="400">



- **Identification:** Recognize the specific risks associated with ML and AI projects, such as data bias, model interpretability, algorithmic transparency, security vulnerabilities, and unintended consequences of AI behavior.
- **Analysis:** Evaluate the nature and magnitude of identified risks. This involves understanding the probability of occurrences (e.g., data breaches, model failures, bias) and their impact on various stakeholders, including users, the organization, and society at large.
- **Evaluation:** Prioritize risks based on their severity and likelihood. This helps in determining which risks need immediate attention and which can be monitored over time.

<br>
<br>
<br>
<br>
<br>


<br>
<br>
<br>
<br>
<br>



# References

References on the OWASP AI guide (a project of which this document is part):
Expand Down Expand Up @@ -1657,5 +1746,7 @@ Misc.:
- See General controls
- MODELINPUTCONFIDENTIALITY (runtime appsec)



- [References](https://github.com/OWASP/www-project-ai-security-and-privacy-guide/blob/main/owaspaiexchange.md#references)
- [Expanded Table of contents](https://github.com/OWASP/www-project-ai-security-and-privacy-guide/blob/main/owaspaiexchange.md#expanded-table-of-contents)