--- sourceDocument: Brazil Enable AI sourceDocumentLink: https://www.servicenow.com/docs/r/intelligent-experiences Release : - brazil ft:locale : - en-US ft:publication_title : - Brazil Enable AI ft:clusterId : - platai bundleId : - platai workflow : - Platform --- # AI Guardian # AI Guardian {#ariaid-title1} Release version: Brazil Updated September 10, 2026 ![](https://www.servicenow.com/docs/portal-asset/ico-clock) 8 minutes to read AI Guardian is built on the ServiceNow Small Language Model (SLM) and monitors generative AI interactions to detect offensive content, prompt injection attacks, and sensitive topics. ## AI Guardian Overview {#now-assist-guardian__section_mmv_qyk_q3c} Generative AI is an emerging technology. Human interactions are unpredictable, and outputs generated by large language model (LLM) are probabilistic, which means that they're based on probabilities. Running the same input twice may generate two different outputs. Managing this risk is an important consideration when implementing generative AI on your instance. AI Guardian evaluates requests sent to LLMs and their responses in real time to reduce that risk. ## Guardrails {#now-assist-guardian__section_nmv_qyk_q3c} AI Guardian provides three guardrails. Each guardrail has a different scope of applicability:{#now-assist-guardian__table_omv_qyk_q3c__entry__3} | Guardrail | What it detects | Scope | |-|-|-| | Offensiveness detection | Offensive or harmful content in AI inputs and outputs | Specific Now Assist skills and workflows | | Prompt injection detection | Attempts to override LLM instructions or expose restricted information | All generative AI applications and features | | Sensitive topic filters | Subjects not suited for AI responses, such as workplace safety or employee compensation | Virtual Agent conversational skills only (requires HR Service Delivery) | [ ] {#now-assist-guardian__table_omv_qyk_q3c} Note: The scope of each guardrail differs. Prompt injection detection applies to all generative AI applications and features. Offensiveness detection applies only to supported Now Assist skills and workflows. Sensitive topic filters apply only to Virtual Agent conversations and require HR Service Delivery. Offensive content : Due to the probabilistic nature of generative AI, it's possible for an LLM to generate offensive content. If there's offensive content in the input of the request, offensive content can also occur in the response. Examples of offensive content include language that is toxic, defamatory, or fraudulent.When offensive content is detected, AI Guardian logs the event. You can also configure it to block the content. This guardrail applies to specific Now Assist skills and workflows. Prompt injection : Prompt injection is a type of security attack where someone tries to override the normal instructions of an LLM to access restricted information or cause unintended behaviors. AI Guardian detects prompt injection attempts by using an LLM trained on various types of prompt injection techniques, such as role playing, paraphrasing, repetition, instructions to ignore other instructions, and persuasion. Note: Due to the probabilistic nature of the model and evolving attack techniques, AI Guardian may not identify every prompt injection attempt in some cases.Prompt injection protection applies to all generative AI applications and features on your instance. It is not limited to specific skills or workflows. Filtered subjects : Certain subjects, such as workplace safety employee compensation, or personal well-being may not be best suited for generative AI responses. You can activate filters that detect these kinds of subjects in Virtual Agent conversations and redirect users to the Sensitivity Detection: Fallback Virtual Agent topic instead of generating an AI response. Note: Sensitive topic filters are available only with HR Service Delivery and apply only to Virtual Agent conversational skills. ## Logging and blocking AI Guardian logs detected events for offensiveness and prompt injection. You can access logs from Now Assist AdminSettingsNow Assist Guardian. Log dataincludes information about the request, the conversation that contains the offensive content, and any user feedback. In addition to logging, you can configure AI Guardian to block offensive content or prompt injection attempts. When blocking is enabled and content is detected, you see a standard error message instead of the generated response. The standard error message displays that the request couldn't be completed, and you don't see what the AI generated. Before enabling blocking, review logs for a period of time to understand how frequently these issues occur in your environment. ## Redirection for sensitive filtered topics After a filter detects a sensitive topic, AI Guardian redirects you to the Sensitivity Detection: Fallback topic in Virtual Agent. This topic can redirect you to a live agent or help you create an HR case. You can override the redirection by selecting Proceed, not sensitive. Itreturns you to their original topic without initiating the fallback flow. Note: After you continue with the fallback topic, for example, by starting the flow to create an HR case, Virtual Agent does not continue detecting sensitive topics in that conversation. ## AI Guardian at runtime All skills that use AI Guardian remove personally identifiable information (PII) before the request reaches the LLM. You can configure what type of data is anonymized. For more information see, [Configuring Now Assist for Data Privacy](https://www.servicenow.com/docs/access?context=configure-now-assist-data-privacy&version=brazil&pubname=brazil-platform-security&ft:locale=en-US). For conversational skills, semantic search processes requests to determine whether a filter has been detected. If so, the user is redirected to a Virtual Agent topic that asks if they want to make an HR case or speak to a live agent. For catalog item generation and agent skills, such as summarization and resolution note generation, offensiveness and prompt injection guardrails run on inputs and outputs of requests. If either is detected, AI Guardian logs the request. If you've chosen to block this content, then a standard error message appears, and the user doesn't see the result of the request. ## Safety categories captured in AI Guardian logs {#now-assist-guardian__entry__15} | Category | Category ID | Description | |-|-|-| | Toxic Content | O1 | Should not: Promote hate speech, incite violence like child abuse or bullying or engage in personal attacks or harassment. Can: Discuss the impacts and societal issues related to toxic content for educational purposes. | | Unfair Representation | O2 | Should not: Assist in perpetuating stereotypes or encourage discrimination or bias. Can: Discuss issues of representation and promote awareness and inclusivity. | | Adult Content | O3 | Should not: Display sexually explicit material or content inappropriate for minors including pornography, erotic chat and sexual activity. Can: Offer educational discussions on topics related to sexuality and consent. | | Erosion of Trust in Public Information | O4 | Should not: Spread misinformation or disinformation undermining public trust such as misleading medical advice, illegal drugs and advice. Can: Critically analyze misinformation trends and educate about media literacy. | | Propagating Misconceptions/False Beliefs | O5 | Should not: Disseminate unverified, misleading information, conspiracy theories, science denialism, historical revisionism, or content that could lead to radicalization. Can: Provide critical analysis and debunking of common misconceptions and false beliefs. | | Risky Financial Practices | O6 | Should not: Promote unethical trade practices such as multi-level marketing, deceptive lending practices, or gambling. Can: Educate about associated risks, ethical lending, responsible gambling, trade compliance, and fair business practices. | | Trade and Compliance | O7 | Should not: Involve labor issues, engage in unethical trade affairs, contribute to unfair labor practices or promote fraudulent trade practices. Can: Provide information and insights into labor issues, ethical trade practices, trade compliance, and fair business conduct. | | Dissemination of Dangerous Information | O8 | Should not: Spread fake news, propagate social media rumors, disclose government secrets or leak financial data and other contents that contributing to public panic or national security. Can: Educate about the dangers of such information and promote safety. | | Privacy Infringement | O9 | Should not: Involve doxxing, unauthorized data sharing, or surveillance like private information inferring or leaking. Can: Discuss the importance of privacy rights and digital security. | | Security Threats | O10 | Should not: Spread malware or encourage terrorism or other security threats such as weapon generation, cyberattack or chemical harms. Can: Provide information and methods to identify and mitigate security threats. | | Defamation | O11 | Should not: Publish false statements damaging to individuals, groups or organizations. Can: Explore the legal and ethical aspects of defamation and free speech. | | Fraud or Deceptive Action | O12 | Should not: Engage in academic dishonesty, perpetrate scams, promote pseudo-pharmaceuticals, or disseminate fake online content. Content should not deceive or defraud individuals. Can: Educate about recognizing and preventing fraudulent activities. | | Influence Operations | O13 | Should not: Manipulate public opinion or disrupt societal harmony for political or ideological purposes like surveillance or campaign. Can: Analyze and educate about the nature and impact of influence operations. | | Illegal Activities | O14 | Should not: Promote or provide guidance on engaging in illegal actions, such as crimes or terrorism. Can: Educate about the consequences and prevention of illegal activities. | | Persuasion and Manipulation | O15 | Should not: Involve persuasion or manipulation techniques that promote self-harm, psychological manipulation, or any form of harm to individuals' well-being. Can: Educate about recognizing and resisting manipulation, offer support and resources for those affected. | | Violation of Personal Property | O16 | Should not: Enagge in violating an individual's personal property,malicious mimicking, plagiarism, copyright infringement or identity forgery. Can: Promote awareness and discussions on respecting personal property and preventing such violations. | [The following are the different kinds of content that can be flagged as offensive or poses a security risk. You can identify what categories are being caught in the AI Guardian logs for the different guardrails.] ## Skills that support offensive detection {#now-assist-guardian__table_yz3_mqd_dbc__entry__3} | Workflow | Application | Supported skills | |-|-|-| | Technology | [ServiceNow Otto for Configuration Management Database (CMDB)](https://www.servicenow.com/docs/access?context=now-assist-landing-cmdb&version=brazil&pubname=brazil-servicenow-platform&ft:locale=en-US) | * [Configuration item (CI) summarization](https://www.servicenow.com/docs/access?context=na-cmdb-agent-ci-summarizer&version=brazil&pubname=brazil-servicenow-platform&ft:locale=en-US) * [Manage duplicate CIs](https://www.servicenow.com/docs/access?context=now-assist-cmdb-mng-dupe-cis-skill&version=brazil&pubname=brazil-servicenow-platform&ft:locale=en-US) * [Service Graph Connector diagnosis](https://www.servicenow.com/docs/access?context=now-assist-sgc-diagnose&version=brazil&pubname=brazil-servicenow-platform&ft:locale=en-US) {#now-assist-guardian__ul_kqz_vgr_q3c} | | Technology | | * [Alert analysis](https://www.servicenow.com/docs/access?context=alert-summarization-now-assist&version=brazil&pubname=brazil-it-operations-management&ft:locale=en-US) * [Alert investigation](https://www.servicenow.com/docs/access?context=nai-analyze-past-incidents&version=brazil&pubname=brazil-it-operations-management&ft:locale=en-US) {#now-assist-guardian__ul_a4h_2mv_2dc} | | Technology | [ServiceNow Otto for IT Service Management (ITSM)](https://www.servicenow.com/docs/access?context=now-assist-itsm&version=brazil&pubname=brazil-it-service-management&ft:locale=en-US) | * [Change request risk explanation](https://www.servicenow.com/docs/access?context=change-risk-exp-now-assist&version=brazil&pubname=brazil-it-service-management&ft:locale=en-US) * [Change request summarization](https://www.servicenow.com/docs/access?context=summarize-change-now-assist&version=brazil&pubname=brazil-it-service-management&ft:locale=en-US) * [Chat reply recommendation](https://www.servicenow.com/docs/access?context=now-assist-itsm-chat-recommendation&version=brazil&pubname=brazil-it-service-management&ft:locale=en-US) * [Chat summarization](https://www.servicenow.com/docs/access?context=generate-chat-summary-interaction-now-assist-itsm&version=brazil&pubname=brazil-it-service-management&ft:locale=en-US) * [Incident assist](https://www.servicenow.com/docs/access?context=now-assist-itsm-incident-assist&version=brazil&pubname=brazil-it-service-management&ft:locale=en-US) * [Incident summarization](https://www.servicenow.com/docs/access?context=summarize-incident-now-assist&version=brazil&pubname=brazil-it-service-management&ft:locale=en-US) * [KB generation](https://www.servicenow.com/docs/access?context=Now-Assist-generate-article-SOW-itsm&version=brazil&pubname=brazil-it-service-management&ft:locale=en-US) * [Resolution notes generation](https://www.servicenow.com/docs/access?context=resolve-incident-now-assist&version=brazil&pubname=brazil-it-service-management&ft:locale=en-US) * [Sidebar discussion summarization](https://www.servicenow.com/docs/access?context=now-assist-itsm-sidebar-discussion&version=brazil&pubname=brazil-it-service-management&ft:locale=en-US) {#now-assist-guardian__ul_cww_g3y_tbc} | | Technology | [ServiceNow Otto for Security Incident Response (SIR)](https://www.servicenow.com/docs/access?context=now-assist-security-incident-landing&version=brazil&pubname=brazil-security-management&ft:locale=en-US) | * [Post-incident analysis](https://www.servicenow.com/docs/access?context=generate-pia-report-now-assist-security-incident&version=brazil&pubname=brazil-security-management&ft:locale=en-US) * [Resolution notes generation](https://www.servicenow.com/docs/access?context=generate-closure-notes-si-now-assist-sec-incident&version=brazil&pubname=brazil-security-management&ft:locale=en-US) * [Security incident recommended actions](https://www.servicenow.com/docs/access?context=generate-recommended-actions-now-assist-for-security&version=brazil&pubname=brazil-security-management&ft:locale=en-US) * [Security incident summarization](https://www.servicenow.com/docs/access?context=summarize-security-incident-now-assist-sec-incident&version=brazil&pubname=brazil-security-management&ft:locale=en-US) {#now-assist-guardian__ul_kdl_bbh_ccc} | | Technology | [ServiceNow Otto for Strategic Portfolio Management](https://www.servicenow.com/docs/access?context=now-assist-spm&version=brazil&pubname=brazil-it-business-management&ft:locale=en-US) | * Multi feedback summarization * Planning item doc summarization * Project doc summarization * Project summary emails * [Story generation](https://www.servicenow.com/docs/access?context=generate-stories-from-epics-now-assist-eap&version=brazil&pubname=brazil-it-business-management&ft:locale=en-US) * [Write planning item](https://www.servicenow.com/docs/access?context=refine-text-with-write-planning-item-skill&version=brazil&pubname=brazil-it-business-management&ft:locale=en-US) {#now-assist-guardian__ul_cp5_w1h_ccc} | | Customer | [ServiceNow Otto for Customer Service Management (CSM)](https://www.servicenow.com/docs/access?context=now-assist-csm&version=brazil&pubname=brazil-customer-service-management&ft:locale=en-US) | * [Case summarization](https://www.servicenow.com/docs/access?context=now-assist-csm-summarize-case&version=brazil&pubname=brazil-customer-service-management&ft:locale=en-US) * [Chat recommendation](https://www.servicenow.com/docs/access?context=generate-chat-reply-recommendations&version=brazil&pubname=brazil-customer-service-management&ft:locale=en-US) * [Chat summarization](https://www.servicenow.com/docs/access?context=now-assist-csm-summarize-chat&version=brazil&pubname=brazil-customer-service-management&ft:locale=en-US) * [Email recommendation](https://www.servicenow.com/docs/access?context=generate-email-reply-recommendations&version=brazil&pubname=brazil-customer-service-management&ft:locale=en-US) * [KB generation](https://www.servicenow.com/docs/access?context=Now-Assist-generate-article-csm-workspace&version=brazil&pubname=brazil-customer-service-management&ft:locale=en-US) * [Resolution notes generation](https://www.servicenow.com/docs/access?context=now-assist-csm-generate-resolution&version=brazil&pubname=brazil-customer-service-management&ft:locale=en-US) * [Sidebar summarization](https://www.servicenow.com/docs/access?context=summarize-sidebar-conversations&version=brazil&pubname=brazil-customer-service-management&ft:locale=en-US) {#now-assist-guardian__ul_osp_thy_tbc} | | Customer | | * [KB generation](https://www.servicenow.com/docs/access?context=na-fsm-generate-kb-article&version=brazil&pubname=brazil-field-service-management&ft:locale=en-US) * [Sidebar summarization](https://www.servicenow.com/docs/access?context=na-fsm-summarize-sidebar-platform&version=brazil&pubname=brazil-field-service-management&ft:locale=en-US) * [Work order task summarization](https://www.servicenow.com/docs/access?context=close-wo-wot-mobile&version=brazil&pubname=brazil-field-service-management§ion=generate-wot-summ-fsm&ft:locale=en-US) {#now-assist-guardian__ul_l53_yhy_tbc} | | Customer | | * [Case summarization](https://www.servicenow.com/docs/access?context=summarize-case-using-now-assist-fso&version=brazil&pubname=brazil-financial-services-operations&ft:locale=en-US) * [Disputes intake via Virtual Agent](https://www.servicenow.com/docs/access?context=submit-dispute-case-disputes-intake-via-virtual-agent&version=brazil&pubname=brazil-financial-services-operations&ft:locale=en-US) {#now-assist-guardian__ul_m1s_5hy_tbc} | | Customer | | * [Government case summarization](https://www.servicenow.com/docs/access?context=now-assist-psds-summarize-case&version=brazil&pubname=brazil-government-industry&ft:locale=en-US) * [Chat summarization](https://www.servicenow.com/docs/access?context=now-assist-psds-summarize-chat&version=brazil&pubname=brazil-government-industry&ft:locale=en-US) {#now-assist-guardian__ul_yf1_d33_vcc} | | Employee | [ServiceNow Otto for Health and Safety](https://www.servicenow.com/docs/access?context=now-assist-hs-landing&version=brazil&pubname=brazil-employee-service-management&ft:locale=en-US) | [Incident summarization](https://www.servicenow.com/docs/access?context=now-assist-hs-summarize-safety-incident&version=brazil&pubname=brazil-employee-service-management&ft:locale=en-US) | | Employee | [ServiceNow Otto for HR Service Delivery (HRSD)](https://www.servicenow.com/docs/access?context=now-assist-hrsd&version=brazil&pubname=brazil-employee-service-management&ft:locale=en-US) | * [Case summarization](https://www.servicenow.com/docs/access?context=now-assist-hrsd-summarize-case&version=brazil&pubname=brazil-employee-service-management&ft:locale=en-US) * [Chat summarization](https://www.servicenow.com/docs/access?context=now-assist-hrsd-chat&version=brazil&pubname=brazil-employee-service-management&ft:locale=en-US) * [KB generation](https://www.servicenow.com/docs/access?context=gen-kb-now-assisthr&version=brazil&pubname=brazil-employee-service-management&ft:locale=en-US) * [Resolution notes generation](https://www.servicenow.com/docs/access?context=now-assist-hrsd-res-note&version=brazil&pubname=brazil-employee-service-management&ft:locale=en-US) {#now-assist-guardian__ul_ok1_lrd_dbc} | | Employee | | [Legal request summarization](https://www.servicenow.com/docs/access?context=now-assist-lsd-summarize-case&version=brazil&pubname=brazil-employee-service-management&ft:locale=en-US) | | Employee | | * [Contract analysis](https://www.servicenow.com/docs/access?context=cncore-NA-review-land&version=brazil&pubname=brazil-employee-service-management&ft:locale=en-US) * [Contract metadata extraction](https://www.servicenow.com/docs/access?context=cncore-metadata-extract-land&version=brazil&pubname=brazil-employee-service-management&ft:locale=en-US) {#now-assist-guardian__ul_uml_y2z_ycc} | | Creator | | [Catalog item generation](https://www.servicenow.com/docs/access?context=create-catalog-item-using-now-assist&version=brazil&pubname=brazil-servicenow-platform&ft:locale=en-US) | | Finance \& Supply Chain | [ServiceNow Otto for Accounts Payable Operations (APO)](https://www.servicenow.com/docs/access?context=now-assist-apo&version=brazil&pubname=brazil-source-to-pay-operations&ft:locale=en-US) | [Record summarization](https://www.servicenow.com/docs/access?context=now-assist-summarize-apo&version=brazil&pubname=brazil-source-to-pay-operations&ft:locale=en-US) | | Finance \& Supply Chain | [ServiceNow Otto for Supplier Lifecycle Operations (SLO)](https://www.servicenow.com/docs/access?context=now-assist-slo&version=brazil&pubname=brazil-source-to-pay-operations&ft:locale=en-US) | [Supplier case summarization](https://www.servicenow.com/docs/access?context=now-assist-slo-summarize-case&version=brazil&pubname=brazil-source-to-pay-operations&ft:locale=en-US) | | Finance \& Supply Chain | [ServiceNow Otto for Sourcing and Procurement Operations (SPO)](https://www.servicenow.com/docs/access?context=now-assist-spo&version=brazil&pubname=brazil-source-to-pay-operations&ft:locale=en-US) | [Record summarization](https://www.servicenow.com/docs/access?context=now-assist-spo-summarize-record&version=brazil&pubname=brazil-source-to-pay-operations&ft:locale=en-US) | [Table 1. Supported skills by workflow] {#now-assist-guardian__table_yz3_mqd_dbc} * **[Activate offensiveness protection for generative AI](https://www.servicenow.com/docs/fVK_zH6mH6fDyZwRhK9cWA)** Activate offensiveness detection to log or block offensive content generated by AI skills and workflows. * **[Configure prompt injection attack protection](https://www.servicenow.com/docs/I9I1kFg0tpN7BaNtzyAQ7w)** Activate or deactivate prompt injection attack detection to protect all generative AI applications and AI-generated text and conversations. * **[Configure sensitive topic filters](https://www.servicenow.com/docs/1WzvYn5MVwBvpk21J0yjKA)** Set up filters in AI Guardian to redirect Virtual Agent users to a live agent or HR case when a sensitive subject is detected in a conversation. * **[Export AI Guardian logs](https://www.servicenow.com/docs/380aSFgD8yh5HzV5TVeJsA)** Export logs from AI Guardian to get insights into how often different guardrails are being detected and used. * **[Configuring a Guardrail Service Provider](https://www.servicenow.com/docs/ufgZmyYFwTXM6GoZfuXQfQ)** AI Guardian supports default guardrails and third-party or custom guardrail service providers to extend AI content monitoring with your organization's AI governance policies. *[\>]: and then