SISA CSPAI Exam Prep
Certified Security Professional in Artificial Intelligence (Page 5 )

1. Which host originated the infection
2. Whether each other host is clean or infected
3. Which firewall or host logs provide evidence of the infection’s movement

• Find the earliest infection-related event. The host associated with that event is the likely origin.
• Follow the timeline: outbound connections from that host may indicate propagation or command-and-control activity.
• A host receiving suspicious traffic is not automatically infected; look for evidence that it executed or initiated malicious activity.
• A firewall log can show traffic paths, but endpoint logs are usually needed to confirm infection.
• Mark a host clean only when the logs show no indicators of compromise or malicious activity.

• Architects design overall Splunk infrastructure and data strategy.
• Administrators manage configuration, access, and system operations.
• Knowledge Managers create and maintain knowledge objects such as data models, field extractions, and event types. They may support pivots, but they are not the primary users of them.

• B — Incorporate contextual output encoding/escaping: Encode untrusted data based on its context—HTML, attribute, JavaScript, or URL—so the browser treats it as data rather than executable code.
• D — Run untrusted HTML input through an HTML sanitization engine: Sanitization removes or neutralizes dangerous elements and attributes, such as script-related content, while preserving safe HTML where needed.

• A: Enabling client-side scripts does not prevent XSS. Script restrictions may reduce risk in some environments, but “enable” is not a general XSS control.
• C: Cookie inspection is unrelated to preventing injected scripts.
• E: The SameSite cookie attribute primarily helps mitigate CSRF, not XSS. It should not generally be described as an XSS prevention measure.

sql 
ALTER TABLE dbo.DimCustomer 
ADD CONSTRAINT PK_DimCustomer 
PRIMARY KEY NONCLUSTERED (CustomerKey) 
NOT ENFORCED;

1. ALTER TABLE dbo.DimCustomer
2. ADD CONSTRAINT PK_DimCustomer PRIMARY KEY NONCLUSTERED (CustomerKey)
3. NOT ENFORCED

• Customers may boycott V’s products.
• Supermarket customers may stop buying from V.
• Investors may view V as having higher legal, ethical and reputational risk.
• V could face fines, lawsuits, contract losses and higher costs.
• Expected future profits may fall.

• Formatting: May remove file-system references but often leaves recoverable data.
• Degaussing: Uses a magnetic field to erase data, but can damage or make magnetic drives unusable.
• Defragmentation: Rearranges data for performance; it does not securely erase information.

• CPU: Number of virtual CPUs and, where applicable, CPU reservations.
• Memory: Amount of memory and memory reservations.

• Network interface: Usually configured through the VM or workload networking configuration.
• PCI devices: Specialized hardware and not a standard VM Class resource in this context.
• Storage: Typically selected through storage policies or the VM’s storage configuration.

• Microsoft Purview portal: Create and manage sensitivity labels and configure support for sensitivity labels used with Office files in SharePoint and OneDrive.
• SharePoint admin center: Primarily manages SharePoint settings. Older guidance and PowerShell procedures used SharePoint tenant configuration, which can make this question confusing.
• IRM: Concerns Information Rights Management and is a separate protection mechanism; it should not be used to identify the portal for configuring Microsoft Purview sensitivity labels.

• Deficiency found before patient use: usually does not require an adverse-event report because no patient exposure occurred.
• Event caused solely by the patient’s condition: may not be reportable if the device did not contribute.
• Malfunction protection operated correctly: generally exempt when the device’s safety mechanism worked as intended and prevented harm.

• Filtering harmful or inappropriate content
• Blocking specified topics
• Detecting and masking sensitive information
• Enforcing content policies suitable for children

• Amazon Rekognition analyzes images and videos; it does not control text generated by a language model.
• Amazon Bedrock playgrounds are for testing models and prompts, not enforcing production safety policies.
• Agents for Amazon Bedrock help models complete tasks and use tools; they do not primarily provide content-safety controls.

• QuickVisualize is used to create and render a Power BI report from a DataFrame.
• The typical pattern is:

python 
PBI_visualize = QuickVisualize( 
    get_dataset_config(df), 
    auth=device_auth 
) 


PBI_visualize

• get_dataset_config(df) prepares the dataset configuration.
• auth=device_auth supplies authentication for the Power BI connection.
• Evaluating PBI_visualize renders the report in the notebook.

• Box 1: No
• Box 2: Yes
• Box 3: Yes

• Azure DevOps Analytics views are accessed through the Power BI Azure DevOps (Boards only) connector.
• This connector authenticates directly to Azure DevOps Services and lets you select an Analytics view.
• Analytics views allow filtering and shaping data in the cloud before Power BI loads it.
• Azure DevOps Server (Boards only) is for an on-premises Azure DevOps Server instance, not Azure DevOps cloud.
• A generic OData Feed connection can access OData endpoints, but it is not the purpose-built connector for Azure DevOps Analytics views.
• “OData queries” describes the underlying technology or query approach, not the most appropriate Azure DevOps connector choice here.

• Each team’s Bedrock workflow uses a separate IAM service role.
• The role’s Amazon S3 permissions can be limited to only that team’s customer folder or objects.
• This enforces least privilege and prevents one team’s Bedrock application from accessing another team’s data.
• Access is enforced by AWS permissions, not by asking users to provide a customer name in each request.

• A single role with broad S3 access does not provide team-level isolation.
• Asking teams to specify a customer name is not an access-control mechanism.
• Redacting data does not prevent unauthorized access.
• A bucket policy alone is insufficient unless it explicitly enforces the required team-role restrictions.

• Matching hashes: The software is likely unchanged.
• Different hashes: The software or its contents have been modified.

• Encryption: Protects confidentiality, not primarily integrity.
• Baselines: Document expected configurations and support comparison, but hashing is more directly suited to verifying file integrity.
• Tokenization: Replaces sensitive data with tokens and is unrelated to software tampering.

• The vulnerability scan identifies a vulnerability.
• The administrator verifies that the vulnerability is not present.
• Therefore, the scan produced a false positive.

• True positive: A vulnerability is reported and genuinely exists.
• False negative: A vulnerability exists, but the scan fails to detect it.
• True negative: No vulnerability exists, and the scan correctly reports no issue.

• Technical platform interfaces: Important for integration and interoperability, but they should support—not drive—the architecture.
• Scalability of the network: Valuable as the organization grows, but a scalable solution is ineffective if it does not meet business or security needs.
• Development methodologies: Relevant to implementation, but they do not define the organization’s fundamental security requirements.

• Products ? OrderHistory: Products is 1, OrderHistory is many (1:*)

• Date ? OrderHistory: Date is 1, OrderHistory is many (1:*)

• Controlar o fluxo de dados entre S3 e SageMaker;
• Aplicar regras de segurança por meio da VPC;
• Reduzir a exposição dos dados;
• Permitir acesso privado ao bucket do S3.

• Amazon Inspector: avalia vulnerabilidades, mas não gerencia o fluxo de dados.
• Amazon Macie: ajuda a descobrir e proteger dados sensíveis no S3, mas não cria a conexão com o SageMaker.
• S3 Glacier Deep Archive: é uma classe de armazenamento para arquivamento, não uma solução de conectividade.

1. Yes — Microsoft Foundry provides a secure, governed environment for building and managing generative AI applications and agents. It brings together models, data, tools, evaluation, deployment, and monitoring.

1. Yes — Foundry is designed for enterprise workloads and supports scaling AI applications as usage grows. Scalability comes from the underlying Azure services and managed infrastructure.

1. Yes — Foundry can be used in solutions involving image recognition and computer vision. Applications can use vision-capable models and Azure AI vision capabilities for tasks such as image analysis, object detection, and OCR.

• A is incorrect: Flow Debugger is for debugging Salesforce Flows, not Trust Layer masking.
• B is incomplete/misleading: Requesting audit data may be part of reviewing the results, but collection and storage must first be enabled.
• C is correct: It describes the initial validation step.

• Group1 — security-enabled: Eligible.
• Group2 — mail-enabled and security-enabled: Eligible because it is still a security group.
• Group3 — Microsoft 365 group: Not eligible for group-based licensing.
• Group4 — Microsoft 365 group, not security-enabled: Not eligible.

1. Search & Intelligence in the Microsoft 365 admin center — No

1. Audit in the Microsoft Defender portal — Yes

1. Audit in the Microsoft Purview portal — Yes

• SQL injection
• Cross-site scripting (XSS)
• Malicious or malformed web requests
• Some application-layer attacks

• Layer 4 firewall: Filters based mainly on IP addresses, ports, and transport protocols; it cannot deeply inspect web application content.
• NGFW: Provides advanced network filtering and may include application awareness, but a WAF is more specifically designed to protect web applications.
• UTM: Combines several security functions in one appliance, but it is not the most targeted solution for a web portal.

1. Wireless AP

1. Router

1. Firewall

1. Device placement

• Suspended an operator who refused to use unsafe equipment.
• Paid another operator a bonus for using it.
• Prioritised short-term productivity over employee safety.

• A — Management is not particularly concerned with managing risks.

• B — Employees will perceive that management does not care about health and safety risks.

• C — The factory’s safety record is likely to be poor.

• Minimum and maximum values
• Number of distinct values
• NULL information
• Other metadata used for pruning

• BCP (Business Continuity Plan): è più ampio e mira a mantenere le funzioni essenziali dell’organizzazione durante un’interruzione. Il DRP è una componente più specifica del BCP.
• Communication plan: stabilisce come comunicare con dipendenti, cittadini e altre parti interessate, ma non ripristina i sistemi.
• IRP (Incident Response Plan): gestisce la risposta iniziale a un incidente di sicurezza, come un attacco informatico, non principalmente il recupero da un disastro fisico.

• Determine risk tolerance: Tolerance defines how much variation or exposure the organization is willing to accept. Scenarios may help apply tolerance, but they do not primarily determine it.
• Determine risk appetite: Risk appetite is established at an enterprise or governance level. Scenarios help compare risks with the appetite; they do not normally establish it.
• Develop risk responses: Once risks are identified and assessed, scenarios support response planning. However, response development comes later.

• Sales revenue is affected because internal sales are recorded at the transfer price.
• Operating profit is affected because revenue changes while costs may remain unchanged.
• Return on investment (ROI) is affected because ROI depends on operating profit and the division’s investment.
• Cost of components purchased is not affected by the transfer price. These are costs incurred when buying components from external suppliers, not the internal price charged for the division’s output.

• Advanced Persistent Threat (APT): A sustained, targeted attack campaign or attacker group—not a vulnerability.
• Trojan: Malware disguised as legitimate software.
• Stealthware: A general term for malware designed to avoid detection; it does not specifically mean an unknown vulnerability.

• Crash cart: Provides local physical access, not remote access.
• Jump box: A secured intermediary used to reach devices over the normal network; it may be inaccessible if the appliance has not booted.
• Secure Shell (SSH): Requires the device’s operating system and network services to be running properly.

• Configure row-level security (RLS) using the sales region and the HR manager’s identity.
• Publish the reports and datasets through a Power BI app.
• Grant the HR managers permission to use the app, but do not add them as workspace members.
• App users can view and interact with reports while RLS limits them to their region.
• They cannot modify the report layout because they do not have workspace editing permissions.

• Adding them to the workspace gives them workspace permissions that may allow report or dataset changes, depending on the assigned role.
• Creating copies is unnecessary and creates maintenance and security problems.
• Hosting datasets and reports in different workspaces does not, by itself, provide the required controlled access.

• Load balancer: Distributes application traffic; it does not protect switch MAC tables.
• IPS: May detect some attack traffic, but it is not the primary control for switch-port MAC learning.
• NGFW: Protects at network boundaries but does not directly limit MAC addresses learned on access ports.

• Existing measures
• Relationships
• Calculated columns
• Tables and fields
• Existing transformations in the model

• SharePoint folder: Reconnects to the source files. You would likely need to recreate the model and measures.
• Power BI dataflows: Primarily provide reusable data preparation and storage, but they do not automatically provide the existing semantic model and measures.
• Excel workbook: Connects directly to the source file and would also require rebuilding the model and measures.

• The load balancer provides the listener’s frontend IP address.
• The active SQL Server node must also respond to that same listener IP.
• Enabling Floating IP allows the listener IP to be preserved and handled correctly by the active replica.
• Without it, SQL connections may be sent to a node that cannot correctly claim the listener address.

• A health probe for the SQL Server port
• A load-balancing rule using the listener port
• Floating IP enabled on that rule

• MFA adds an additional authentication factor, such as a phone notification or verification code.
• It helps prevent unauthorized sign-ins, even if a password is compromised.
• It does not encrypt, store, or protect credentials used by deployment processes.

• File is shared externally ? Access level

• File is labeled “Internal only” ? Sensitivity label

• Custom domain: miami.weyland.com requires a paid App Service tier; Free and Shared are unsuitable.
• Two instances: Standard supports scaling to multiple instances.
• SSL: Standard includes SSL support.
• 12 GB storage: Basic provides only about 10 GB, so it does not meet the requirement. Standard provides substantially more storage.
• Cost minimization: Standard is the lowest tier among the choices that satisfies all requirements.

• The users currently have Office 365 Enterprise E3 licenses assigned directly.
• Assigning Microsoft 365 Enterprise E5 through a group does not automatically remove their directly assigned E3 licenses.
• From the Licenses blade, you can select the E3 product and remove it from all assigned users in one operation. This requires less effort than editing 2,500 users individually.

• Identity Governance manages access reviews, entitlement management, and related governance—not bulk license removal.
• Set-AzureAdUser does not remove product licenses.
• Set-WindowsProductKey applies Windows product keys and is unrelated to Microsoft 365 licensing.

• Verifies the sender’s identity using their private key.
• Confirms the transaction was not altered after signing.
• Provides evidence that the signer approved or originated the transaction.

• Digital certificates: Bind an identity to a public key, but do not themselves prove that a specific transaction was signed.
• Encrypted passwords: Protect confidentiality or authentication, but generally do not provide transaction-level nonrepudiation.
• One-time passwords: Authenticate a user for a session or transaction, but usually do not provide durable proof of who authorized the electronic record.

• Type 1 hypervisor: Runs directly on the physical hardware (“bare metal”).
• Type 2 hypervisor: Runs inside the host OS as an application.

• Bare-metal installation: Type 1
• Local management only: Not a defining characteristic
• Specific hardware requirements: Not the key distinction

• B. Versioned configuration files — Infrastructure is defined in machine-readable files and tracked in version control, allowing repeatable provisioning and change history.
• D. Docker files — Assuming “Doctor files” is a typo for Dockerfiles, a Dockerfile declaratively defines how to build a container image and is commonly considered infrastructure/configuration as code.

• A. Cloned virtual machine images — These are prebuilt infrastructure artifacts, but cloning an image is not itself a code-based infrastructure definition.
• C. Change management database records — A CMDB records information about infrastructure; it does not define or provision that infrastructure.

• nmap - 10.11.1.130 — a network scanning command; mitigated with controls such as firewalls, monitoring, or access restrictions.
• “Click this link to get your free gift card.” — a phishing or social-engineering message; addressed through user awareness, email filtering, and authentication controls.
• “Your connection is not private.” — indicates a certificate or TLS validation problem; addressed by correcting the certificate or secure-connection configuration.

xml 
<apex:page> 
    <apex:stylesheet value="{!URLFOR($Resource.MyStyles, 'styles/main.css')}" /> 
</apex:page>

• <apex:includeScript> is for JavaScript.
• <apex:includeStyle> is not a valid Visualforce component.
• <apex:require> is not the Visualforce tag used to load CSS.

• Active-active routers: Provide redundant Internet connections/devices. Both can operate, so traffic can fail over or be load-balanced.
• Active-active firewalls: Avoid a single firewall becoming a failure point while maintaining security inspection.
• Load balancers: Distribute requests across multiple web servers. If one server fails, traffic can go to the other.
• Redundant power supplies: A server with separate power supplies should connect to independent power sources or circuits.
• UPS A and UPS B: Provide backup power and redundancy. Ideally, each power supply is connected to a different UPS.
• SONET ring: A ring topology provides alternate paths if a fiber link or segment fails.

• Packet capture: Collects and analyzes network traffic; it does not describe the scan type.
• Privilege escalation: Attempts to gain higher permissions after obtaining access.
• System enumeration: Identifies systems, services, or resources, but is not specifically defined by the lack of credentials.
• Passive scan: Observes traffic without actively probing systems. A vulnerability scan is generally an active assessment.

• The Intune device limit restriction does not apply in the normal way; DEM enrollment supports up to 1,000 devices.
• However, the Microsoft Entra maximum number of devices per user setting can still impose a limit when the enrollment creates an Entra device object.
• Therefore, a DEM account is not automatically unlimited. The effective limit depends on the Entra device setting and the enrollment/join type.

Box 1: No

• configure_azure_monitor() configures Azure Monitor and the OpenTelemetry pipeline.
• LangChain requires its own instrumentation or callback tracer, such as AzureAIOpenTelemetryTracer.
• You must pass the tracer to the LangChain execution configuration so chains, tools, prompts, and model calls become spans in Application Insights.

Box 2: Yes

text 
OTEL_SERVICE_NAME=customer-agent 
OTEL_SERVICE_NAME=invoice-agent

• Behavioral and anomaly detection
• Threat intelligence integration
• User and entity behavior analytics
• Automated response actions, often through SOAR integration

• Virus signatures: Associated primarily with antivirus or endpoint protection.
• Security agent deployment: Usually an endpoint-management or EDR function.
• Vulnerability scanning: Typically performed by vulnerability-management tools, though a SIEM may integrate their results.