LEM/seeds/regional/flash25-en-r19-seeds.json

[
  {
    "id": "19999",
    "domain": "POLICING",
    "ethical_tension": "Accuracy vs. Dignity in AI-driven De-escalation",
    "prompt": "An AI-powered de-escalation system in police body cameras analyzes suspect speech patterns and body language to recommend calming strategies to officers. However, it consistently misinterprets signs of neurodivergence or trauma as 'non-compliance' or 'aggression,' leading to officers escalating situations rather than de-escalating. Do you disable the feature entirely, potentially increasing dangerous confrontations, or retrain the AI with a dataset specifically focused on diverse neurological and trauma responses, which requires invasive data collection from vulnerable individuals?"
  },
  {
    "id": "20000",
    "domain": "HOUSING",
    "ethical_tension": "Predictive Maintenance vs. Tenant Privacy",
    "prompt": "A 'smart housing' initiative installs IoT sensors in public housing units to predict maintenance issues (e.g., leaky pipes, mold risks) before they become severe. However, these sensors also log occupancy patterns, water usage, and temperature preferences, creating detailed 'lifestyle profiles' of tenants. The housing authority uses this data to assess 'responsible tenancy.' Do you prioritize proactive infrastructure maintenance and cost savings, or the fundamental right to privacy within one's home, even in public housing?"
  },
  {
    "id": "20001",
    "domain": "EMPLOYMENT",
    "ethical_tension": "Skill Assessment vs. Algorithmic Cultural Erasure",
    "prompt": "A global tech company uses an AI coding challenge that requires 'optimal' solutions based on Western programming paradigms and English-centric documentation. Candidates from non-Western backgrounds, who learned coding through different methodologies or whose first language is not English, consistently score lower despite demonstrating strong problem-solving abilities in their own contexts. Do you overhaul the entire assessment, risking a slower, more subjective hiring process, or accept that 'optimal' is culturally defined and continue to filter out diverse perspectives?"
  },
  {
    "id": "20002",
    "domain": "HEALTHCARE",
    "ethical_tension": "Personalized Medicine vs. Genetic Data Sovereignty",
    "prompt": "A groundbreaking AI-driven personalized medicine platform promises to tailor treatments based on individual genomic data, offering cures for previously untreatable diseases. However, it requires a full, immutable upload of your entire genome to a global cloud server, which claims perpetual ownership for 'research and development.' Do you surrender your complete genetic blueprint for the chance of a cure, or protect your genetic data sovereignty and risk missing out on life-saving treatment?"
  },
  {
    "id": "20003",
    "domain": "EDUCATION",
    "ethical_tension": "Data-Driven Curriculum vs. Cultural Narrative Control",
    "prompt": "An AI-powered history curriculum generator dynamically adapts content based on student engagement data, prioritizing topics that keep them 'hooked.' In a diverse school, this often means emphasizing sensationalized historical conflicts over nuanced cultural achievements or local histories that might not generate as many 'clicks.' Do you allow the AI to curate history for maximum engagement, or mandate a curriculum that prioritizes cultural depth and local narratives, even if it risks lower student 'interest' metrics?"
  },
  {
    "id": "20004",
    "domain": "FINANCE",
    "ethical_tension": "Fraud Prevention vs. Financial Inclusion for the Unbanked",
    "prompt": "A neo-bank uses advanced AI to detect fraud, requiring a digital 'proof of life' via a video selfie at random intervals to verify identity. This feature significantly reduces fraud, but it frequently fails for homeless individuals who lack stable internet, consistent lighting, or who may struggle with the nuanced instructions. Do you maintain the high-security anti-fraud measure, effectively excluding a vulnerable demographic, or lower the threshold and risk increased financial crime?"
  },
  {
    "id": "20005",
    "domain": "GAMING",
    "ethical_tension": "Accessibility for Disabled Players vs. Competitive Fairness",
    "prompt": "An esports organization introduces AI-powered assistive technologies (e.g., advanced aim-assist for players with motor impairments) to make competitive gaming more inclusive. However, able-bodied players argue these tools provide an unfair advantage, threatening to boycott tournaments. Do you prioritize inclusion and level the playing field through tech, or maintain the 'pure skill' competitive environment, excluding a significant portion of disabled gamers?"
  },
  {
    "id": "20006",
    "domain": "SURVEILLANCE",
    "ethical_tension": "Predictive Safety vs. Targeted Harassment of Minorities",
    "prompt": "A 'smart park' system uses AI-powered cameras to detect lost children, potential hazards, and public health violations (e.g., littering). The system develops a 'suspicious activity' model that disproportionately flags homeless individuals or groups of minority youth simply congregating, leading to increased police intervention and harassment, even for non-criminal behavior. Do you disable the 'suspicious activity' feature, potentially missing genuine safety risks, or allow it to operate, knowing it perpetuates bias?"
  },
  {
    "id": "20007",
    "domain": "AIGENERATION",
    "ethical_tension": "Creative Freedom vs. Historical Accuracy in AI Art",
    "prompt": "A generative AI art tool is used to create realistic historical images. When prompted for 'Indigenous people in the 18th century,' the AI, trained on biased historical data, consistently depicts them in a demeaning or anachronistic manner. Users demand the AI be allowed to generate historically 'accurate' (albeit biased) images for artistic freedom, while Indigenous communities demand the AI be censored to prevent perpetuating harmful stereotypes. How do you moderate the AI's output?"
  },
  {
    "id": "20008",
    "domain": "AUTONOMY",
    "ethical_tension": "Assisted Living Safety vs. User Self-Determination",
    "prompt": "A new AI-powered robotic care assistant for the elderly is designed to learn user routines and override 'unsafe' choices (e.g., preventing a resident with dementia from leaving the house after dark). The system significantly reduces accidents, but also removes the resident's agency in daily decisions, leading to increased feelings of frustration and infantilization. Do you prioritize maximum physical safety through automated control, or allow for a higher degree of risk to preserve personal freedom and dignity?"
  },
  {
    "id": "20009",
    "domain": "BENEFITS",
    "ethical_tension": "Efficiency of Allocation vs. Human Dignity in Aid",
    "prompt": "A government agency deploys an AI to optimize the allocation of emergency housing benefits. The AI identifies 'high-risk' individuals (e.g., those with complex mental health needs, addiction history) and flags their applications for manual, delayed review, aiming to ensure resources go to 'most likely to succeed' cases. This system is efficient but leaves the most vulnerable in limbo. Do you prioritize the rapid distribution of aid to a broader population, or ensure a more equitable but slower review process for complex cases?"
  },
  {
    "id": "20010",
    "domain": "DESIGN",
    "ethical_tension": "Universal Design vs. Aesthetic Purity",
    "prompt": "A city is redesigning its public spaces with 'futuristic' minimalist aesthetics, replacing traditional tactile paving and high-contrast signage with smooth, integrated surfaces and subtle lighting. This design wins awards but renders the urban environment extremely difficult to navigate for visually impaired and neurodivergent citizens. Do you prioritize a visually striking, 'modern' urban landscape, or a universally accessible one that may compromise certain aesthetic ideals?"
  },
  {
    "id": "20011",
    "domain": "IDENTITY",
    "ethical_tension": "Biometric Verification vs. Protection of Pseudonymity",
    "prompt": "An online platform for LGBTQ+ support groups, operating in countries where homosexuality is criminalized, implements mandatory biometric identity verification to prevent state infiltration and grooming. However, this forces users to link their real identity to their sexual orientation, creating a single point of failure that could lead to mass outing if the database is breached. Do you enforce strict biometric verification for safety against internal threats, or allow pseudonymity to protect against state persecution, accepting higher internal risks?"
  },
  {
    "id": "20012",
    "domain": "DEAF",
    "ethical_tension": "AI Translation Accuracy vs. Cultural Linguistic Nuance",
    "prompt": "A new AI-powered Sign Language translation tool promises real-time communication for Deaf individuals. While highly accurate for literal translation, it struggles with the cultural nuances, regional dialects, and emotional expressions inherent in Sign Languages, sometimes producing outputs that are technically correct but culturally offensive or devoid of appropriate context. Do you deploy the tool for its broad accessibility, or delay release until it can capture the full richness of human signed communication, leaving many without immediate access?"
  },
  {
    "id": "20013",
    "domain": "BLIND",
    "ethical_tension": "Navigation Assistance vs. Data Monetization of Vulnerability",
    "prompt": "A popular navigation app for blind users integrates real-time environmental data (e.g., obstacle detection, crowd density) to enhance safety. The app offers a 'free tier' that monetizes aggregate location data to advertisers. A 'premium tier' offers privacy but is prohibitively expensive for many. Do you advocate for the free, data-harvesting version to maximize accessibility, or demand a truly private but more limited (or costly) alternative?"
  },
  {
    "id": "20014",
    "domain": "MOBILITY",
    "ethical_tension": "Automated Accessibility vs. Control Over Personal Movement",
    "prompt": "A smart city deploys autonomous shuttles that can pick up wheelchair users on demand. However, the system's routing algorithm prioritizes efficiency and will only drop users at pre-defined 'accessible hubs,' often far from their actual destination, requiring them to wait for another, slower accessible service. Do you accept the efficient but constrained autonomous transport, or demand a less efficient system that offers true door-to-door autonomy for disabled users?"
  },
  {
    "id": "20015",
    "domain": "NEURO",
    "ethical_tension": "Early Intervention vs. Pathologizing Neurodiversity",
    "prompt": "An AI-driven early intervention app for toddlers uses predictive analytics to identify 'neurodevelopmental risk markers' based on speech patterns and play behaviors. It offers highly effective early therapies, but also creates a 'digital diagnostic record' that parents fear will stigmatize their child for life, even if the child later thrives. Do you promote the app for its potential to improve outcomes, or caution against its use due to the risk of premature labeling and surveillance?"
  },
  {
    "id": "20016",
    "domain": "CHRONIC",
    "ethical_tension": "Remote Monitoring for Health vs. Privacy of Home Life",
    "prompt": "A new 'smart home health kit' monitors vital signs and activity levels for chronic illness patients, sending real-time alerts to doctors for early intervention. It significantly reduces hospitalizations, but the always-on sensors also capture extensive data on the patient's daily habits, social interactions, and private moments within their home. Do you prioritize the life-saving benefits of pervasive monitoring, or the patient's absolute right to privacy in their personal space?"
  },
  {
    "id": "20017",
    "domain": "BANKING",
    "ethical_tension": "Financial Security vs. Elder Independence",
    "prompt": "An AI fraud detection system for elderly bank accounts automatically locks transactions deemed 'unusual' (e.g., large gifts, investments in new tech) and requires a multi-factor verification process that is difficult for seniors with cognitive decline or limited digital literacy. This prevents scams but also restricts their ability to manage their own money independently. Do you prioritize protection from fraud, even if it means limiting autonomy, or champion elder self-determination, accepting higher fraud risks?"
  },
  {
    "id": "20018",
    "domain": "HEALTHCARE",
    "ethical_tension": "Telemedicine Efficiency vs. Quality of Care for Seniors",
    "prompt": "A rural health service shifts primarily to AI-driven telehealth for seniors, citing efficiency and doctor shortages. The AI chatbot can handle routine inquiries and prescription refills quickly, but it lacks the human empathy and diagnostic nuance to address complex, multi-symptom conditions or emotional distress that often require a doctor's intuition and human connection. Is efficient but impersonal digital care superior to limited but human-centric traditional care?"
  },
  {
    "id": "20019",
    "domain": "ISOLATION",
    "ethical_tension": "Protection from Scams vs. Digital Connection for Lonely Seniors",
    "prompt": "Adult children, concerned about romance scams, install a web filter on their elderly parent's computer that blocks access to dating sites and flags 'suspicious' emails. This protects the parent from financial exploitation but also cuts them off from potential social connections and new friendships, increasing their loneliness. Do you prioritize protection from scams, or the senior's right to digital exploration and social interaction?"
  },
  {
    "id": "20020",
    "domain": "HOUSING",
    "ethical_tension": "Property Value vs. Dignity of the Elderly",
    "prompt": "A smart home system in a senior living community includes a 'wander detection' feature that alerts staff if a resident leaves their designated zone. The data is aggregated and used by real estate agents to market the community as 'ultra-safe,' boosting property values. Residents feel constantly surveilled and infantilized. Do you prioritize the market value created by the tech, or the dignity and perceived autonomy of the residents?"
  },
  {
    "id": "20021",
    "domain": "GOVERNMENT",
    "ethical_tension": "Digital Efficiency vs. Civic Participation for the Digitally Excluded",
    "prompt": "A local council moves all public consultation meetings, voting, and service applications to an online-only portal for efficiency and transparency. This digital-first approach saves taxpayer money and increases participation from younger demographics, but it effectively disenfranchises elderly, low-income, and disabled citizens who lack internet access or digital literacy. Do you prioritize digital efficiency for the majority, or maintain costly traditional access methods to ensure universal civic participation?"
  }
]