Now Assist Guardian - Prompt Injection - How Does it Work?

JustinD56844010
Tera Contributor

3 weeks ago

We are enabling Now Assist in Virtual Agent.

We are also enabling Now Assist Guardian - Prompt Injection.

 

We are struggling with understanding what ServiceNow considers a malicious prompt to be?
If a user using the Virtual Agent in Employee Center asks the Virtual Agent for another employee's home address - is that considered a malicious prompt?

 

The table that stores the logging for the malcious prompts, sn_nowassist_admin_sys_gen_ai_export_data, is not easily readable so it's hard to figure out how 'Prompt Injection' works.

0 Helpfuls
  • 858 Views
2 REPLIES 2

Tanushree Maiti
Tera Patron

3 weeks ago

Hi @JustinD56844010 

 

Check these resources if helps.

Configuring Now Assist Guardian: Offensive Content & Prompt Injection Protection

Protecting Your AI with Now Assist Guardian: Offensive Content Stops Here

ServiceNow Now Assist Guardian: Enterprise Protection That Accelerates Deployment

Now Assist Guardian FAQ

NowAssist Guardian: A Comprehensive Guide for Safer AI Use in ServiceNow

https://www.youtube.com/watch?v=AoEjEvDR088

https://www.youtube.com/watch?v=OFt1XcJ_P18

https://www.youtube.com/watch?v=-MQDRe-o24o

https://www.youtube.com/watch?v=jKBSN26muQI

 

 

 

Please Accept the solution if it assisted you with your question & Mark this response as Helpful.
Regards
Tanushree Maiti
ServiceNow Technical Architect
LinkedIn: https://www.linkedin.com/in/tanushreemaiti
AI Academy: Introducing Now Assist Guardian
AI Academy: Introducing Now Assist Guardian
youtube.com/watch?v=AoEjEvDR088
In this session we cover Now Assist Guardian, our newest feature delivering built-in guardrails for trustworthy AI. Launched in Q4 2024, Now Assist Guardian monitors and blocks offensive content and prompt injection attacks when using Now Assist generative AI products. It also covers sensitive ...
This AI Filter Just Saved Your App: Now Assist Guardian in Action!
This AI Filter Just Saved Your App: Now Assist Guardian in Action!
youtube.com/watch?v=OFt1XcJ_P18
Are you worried about offensive or unsafe content sneaking into your AI experiences? In this quick episode of Did You Know, Willem Zeiler walks us through Now Assist Guardian - a built-in safeguard for AI-generated content on the Now Platform. You'll learn how to: - Enable the offensiveness filter
Building Trust in AI with ServiceNow's Now Assist Guardian
Building Trust in AI with ServiceNow's Now Assist Guardian
youtube.com/watch?v=-MQDRe-o24o
In this episode of TechBytes, AI Governance Product Manager at ServiceNow, Louis Philip Morin, discusses the critical aspects of trust in AI. The conversation delves into the challenges of creating AI experiences free from toxic and malicious language, highlighting the role of Now Assist Guardian
Now Assist Guardian | Overview
Now Assist Guardian | Overview
youtube.com/watch?v=jKBSN26muQI
Now Assist Guardian is a built-in solution for enhancing the safety and reliability of AI powered experiences on the Now platform. Learn what it does, how to configure it, and how to analyze the data it collects. Chapters: 00:00 Now Assist Guardian features 00:56 Offensiveness guardrail 01:21
0 Helpfuls

hugohuyghue
Tera Contributor

3 weeks ago

Hi @JustinD56844010
I do think the resources Tanushree Maiti suggested are very helpful and will be able to assist you.

0 Helpfuls