Basic AI Concepts Related to Cybersecurity

SecAI+ · 34 questions

• A city 311 office pilots a chatbot that drafts resident reply emails from published policy PDFs instead of only assigning ticket categories. Which AI capability best describes that drafting work?
• A water utility trains a model on years of meter telemetry so it can flag anomalous consumption before billing disputes escalate. Which description best matches that approach?
• A tax assessor's office estimates parcel risk scores from property features using statistical learning methods rather than a deep neural network. What distinction should the AI lead emphasize?
• A courts clerk chatbot summarizes long motion filings for docket staff using a modern language model. Which architecture family most directly enables that summarization capability?
• A transit authority analyzes platform camera feeds to estimate crowd density near boarding areas for safety operations. Which AI approach best fits that vision workload?
• A small parks department needs an on-premises assistant for trail FAQs and is weighing a large cloud LLM against a small language model. When is the SLM the better fit?
• A city communications team reviews how generative adversarial networks could be abused to fabricate fake mayor press videos. What risk concept should the team associate with GANs in that scenario?
• A public-health clinic labels prior outbreak cases so a model can predict high-risk ZIP codes for outreach. Which training approach does that labeling enable?
• A police records bureau wants to discover theme groups in incident narratives without predefined category labels. Which learning approach fits that goal?
• An emergency-management dispatcher assistant is tuned so better resource-routing suggestions earn positive feedback rewards. Which training paradigm is being applied?
• County IT fine-tunes a permits FAQ model, then prunes and quantizes it so building-inspector tablets can run it offline. What is the primary goal of pruning and quantization here?
• A housing-authority chatbot must never reveal applicant SSNs even when residents type free-form questions. Which prompt control best enforces that boundary?
• A library helpdesk AI answers overdue-fine questions more consistently after staff paste one example reply into the prompt. Which prompting technique are they using?
• A fire-district incident-response summarizer always uses a locked prompt template and a "safety officer" system role for after-action drafts. What benefit do those controls provide?
• Before training a 311 intent model, city analysts remove duplicate tickets and verify category codes against the service catalog. Which data-security hygiene step is that?
• After an elections chatbot gives a wrong answer, a clerk's office must show which open-data CSV fed the embedding index. What practice enables that investigation?
• A school-district bullying-report classifier overfits the majority "none" label until staff carefully augment and balance rare categories. What problem are they correcting?
• A wastewater plant feeds SCADA numeric tables, JSON work orders, and free-text operator notes into one anomaly assistant. How should the AI team classify those inputs?
• A city wants AI-generated open-meeting summaries marked so public copies can be traced if altered. Which technique directly supports that authenticity goal?
• A permits desk embeds building-code sections so a chatbot can retrieve relevant clauses before answering contractors. What role do those embeddings play?
• County counsel stores ordinance embeddings in a vector database isolated from the public web tier. Why is that design significant for RAG security?
• Public-health staff find silent row corruption in a vaccination-training extract before a scheduled model refresh. What should they prioritize?
• A transit chatbot cites a superseded fare policy because the RAG corpus was never checked after a board vote. Which control was missing?
• An open-data portal tags each dataset with publisher and update stamp before any city LLM may ingest it. What requirement does that tagging satisfy?
• A mayor's office rejects a flashy generative demo that does not map to reducing 311 backlog KPIs. What lifecycle principle is the office applying?
• Emergency management wants to train a flood-damage classifier using resident-submitted photos. Which practice best protects trustworthiness during data collection?
• Housing caseworkers need case notes ready for a fine-tune job. Which data-preparation step best reduces sensitive-data exposure before training starts?
• County IT must choose a model for assessor Q&A about local valuation rules. Which selection approach best matches municipal scope and risk?
• A courts chatbot is nearly ready for clerk staff. Which action best represents a proper model-evaluation gate before go-live?
• Parks IT is rolling out a reservation assistant. Which deployment approach best supports safe municipal validation?
• After winter storms, a road-closure summarizer starts citing outdated detours. Which lifecycle response best addresses the drift?
• 311 agents keep flagging weak chatbot draft replies. Which practice best turns that signal into safer iteration?
• A welfare-eligibility assistant drafts benefit recommendations. Which control best applies human-in-the-loop for high-impact civic decisions?
• An AI tool proposes mutual-aid staging plans during a multi-alarm fire. Which oversight practice is most appropriate before paging resources?