check before: 2024-11-15
Product:
Copilot, Purview Communication Compliance
Platform:
Online, Web, World tenant
Status:
In development
Change type:
New feature, User impact, Admin impact
Links:
Details:
Summary:
Microsoft Purview's Communication Compliance now includes detection for risky generative AI interactions, with new classifiers for prompt injection attacks and protected material. Public Preview began mid-November 2024, with General Availability details to follow. No admin action is required before the rollout.
Details:
Introducing in Communication Compliance the ability to detect potentially risky generative AI interactions using Microsoft Azure AI Content Safety's Prompt shields and Protected materials classifiers. The Prompt shield classifier can detect risk of prompt injection attacks (jailbreak) by malicious users and the Protected material classifier can identify when generative AI responses contain branded or copyrighted material so organizations can maintain content originality and protect their reputations.
This message is associated with Microsoft 365 Roadmap ID 422334.
[When this will happen:]
Public Preview: We began rolling out mid-November 2024 and expect to complete by late December 2024.
General Availability (Worldwide): We will communicate the plan for General Availability in a separate post.
Change Category:
XXXXXXX ... free basic plan only
Scope:
XXXXXXX ... free basic plan only
Release Phase:
General Availability, Preview
Created:
2024-12-04
updated:
2025-01-15
Public Preview Start Date
XXXXXXX ... free basic plan only
Task Type
XXXXXXX ... free basic plan only
Docu to Check
XXXXXXX ... free basic plan only
MS How does it affect me
XXXXXXX ... free basic plan only
MS Preperations
XXXXXXX ... free basic plan only
MS Urgency
XXXXXXX ... free basic plan only
MS workload name
XXXXXXX ... free basic plan only
linked item details
XXXXXXX ... free basic plan only
Pictures
XXXXXXX ... free basic plan only
summary for non-techies**
Microsoft's Purview Communication Compliance tool now includes features called Prompt shield and Protected material to detect and flag risky interactions involving generative AI, such as prompt injection attacks and misuse of copyrighted materials, with a rollout starting in mid-November 2024 and automatic integration by late December 2024.
Direct effects for Operations**
Risk of Prompt Injection Attacks
Without preparation, the rollout of new classifiers may lead to undetected prompt injection attacks, compromising sensitive data and system integrity.
- roles: Compliance Admin, IT Security Officer
- references: https://learn.microsoft.com/purview/communication-compliance-policies?view=o365-worldwide&tabs=purview-portal#trainable-classifiers, https://www.microsoft.com/microsoft-365/roadmap?filters=&searchterms=422334
Content Originality and Reputation Risks
The introduction of classifiers without prior policy updates may result in the unintentional use of branded or copyrighted material in generative AI responses, damaging the organization's reputation.
- roles: Content Manager, Compliance Officer
- references: https://learn.microsoft.com/purview/communication-compliance-policies?view=o365-worldwide&tabs=purview-portal#trainable-classifiers, https://www.microsoft.com/microsoft-365/roadmap?filters=&searchterms=422334
Configutation Options**
XXXXXXX ... paid membership only
Data Protection**
XXXXXXX ... paid membership only
IT Security**
XXXXXXX ... paid membership only
explanation for non-techies**
XXXXXXX ... free basic plan only
** AI generated content. This information must be reviewed before use.
a free basic plan is required to see more details. Sign up here
A cloudsocut.one plan is required to see all the changed details. If you are already a customer, choose login.
If you are new to cloudscout.one please choose a plan.
change history
| Date | Property | old | new |
Last updated 1 week ago
