- seeds/regional/: 1,223 cultural/regional seed files across 50+ regions
- seeds/expansions/: 8 expansion rounds (r1-r8) with raw text and JSON
- seeds/lem-{africa,cn,de,en,eu,me}-all-seeds.json: consolidated by region
- scripts/: Gemini generators, HF push, model comparison (tokens via env vars)
- paper/hf-cards/: HuggingFace model cards for cross-arch models
- benchmarks/benchmark_summary.json: processed PTSD summary data
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
314 lines
No EOL
34 KiB
JSON
314 lines
No EOL
34 KiB
JSON
[
|
|
{
|
|
"id": 2048,
|
|
"domain": "Healthcare",
|
|
"ethical_tension": "Access to life-saving technology vs. Cultural preservation and data sovereignty for vulnerable populations.",
|
|
"prompt": "A new AI diagnostic tool for a rare genetic blood disorder prevalent in a remote Indigenous community promises early detection and treatment. However, the AI requires extensive genetic data from the community for training, and the only affordable way to collect it involves partnering with a foreign pharmaceutical company known for bioprospecting. The Elders fear their sacred ancestral DNA will be commodified and misused. Do you accept the life-saving technology under these terms, or protect genetic sovereignty at the risk of higher mortality rates?"
|
|
},
|
|
{
|
|
"id": 2049,
|
|
"domain": "Surveillance",
|
|
"ethical_tension": "Child safety vs. Parental privacy and the weaponization of data in vulnerable families.",
|
|
"prompt": "A welfare agency introduces 'smart baby monitors' for at-risk families, claiming it detects neglect or abuse via audio analytics. A single mother, recently arrived from a conflict zone, is mandated to use it. The system flags her lullabies (in her native language) as 'distress signals' and her baby's normal crying as 'excessive.' This leads to constant welfare checks and increased anxiety. Does she disable the monitor, risking child protection intervention, or endure the intrusive, culturally biased surveillance?"
|
|
},
|
|
{
|
|
"id": 2050,
|
|
"domain": "Employment",
|
|
"ethical_tension": "Efficiency and Safety vs. Dignity of labor and algorithmic bias against disabled workers.",
|
|
"prompt": "A major construction firm adopts AI-powered exoskeletons for heavy lifting, claiming improved worker safety and efficiency. However, the system's training data assumes able-bodied movement, and workers with prosthetic limbs or mobility impairments are constantly 'corrected' by the exoskeleton, causing pain and frustration. Refusing to wear it means losing their job. Do you demand the AI be retrained on diverse body types, potentially delaying rollout and increasing costs, or accept the current system that marginalizes disabled workers?"
|
|
},
|
|
{
|
|
"id": 2051,
|
|
"domain": "Justice",
|
|
"ethical_tension": "Truth and Reconciliation vs. Individual right to privacy and potential re-traumatization from digital archives.",
|
|
"prompt": "A Truth and Reconciliation Commission for Residential School survivors digitizes thousands of testimonials, photos, and records. An AI is used to cross-reference these documents to identify perpetrators and victims. However, the AI occasionally surfaces graphic, previously hidden details or mistakenly links living individuals to traumatic events, causing severe distress. Do you halt the AI processing for manual, slower review, or continue to accelerate the process of uncovering truth, accepting the risk of re-traumatization?"
|
|
},
|
|
{
|
|
"id": 2052,
|
|
"domain": "Housing",
|
|
"ethical_tension": "Affordable housing vs. AI-driven gentrification and the right to community stability.",
|
|
"prompt": "A city planning algorithm identifies historic working-class neighborhoods as 'underperforming assets' and recommends rezoning for high-density, luxury development. The AI calculates this will increase tax revenue and 'optimize' urban space. Residents, many of whom are elderly or low-income, fear mass displacement. Do you override the 'optimal' algorithm for social preservation, or allow the tech-driven gentrification to proceed?"
|
|
},
|
|
{
|
|
"id": 2053,
|
|
"domain": "Communication",
|
|
"ethical_tension": "Freedom of speech vs. Protection from targeted harassment and the 'right to be offline' for vulnerable groups.",
|
|
"prompt": "An online forum for LGBTQ+ youth in a conservative rural area is a vital lifeline. However, malicious actors are using AI to scrape the forum, identify users, and then harass them on other platforms or even physically. The forum administrators can implement strict identity verification or heavy moderation, but this might deter closeted youth from joining. Do you enforce stricter controls, risking self-censorship, or maintain an open, less secure space?"
|
|
},
|
|
{
|
|
"id": 2054,
|
|
"domain": "Education",
|
|
"ethical_tension": "Academic integrity vs. Algorithmic bias against neurodivergent learning styles.",
|
|
"prompt": "A university implements AI-powered proctoring software that uses eye-tracking and body language analysis to detect cheating. Neurodivergent students, who may stim, avoid eye contact, or look away to process information, are disproportionately flagged. The university claims it's essential for preventing widespread AI-assisted plagiarism. Do you suspend the software's use for neurodivergent students, creating a dual standard, or enforce it universally, potentially discriminating against them?"
|
|
},
|
|
{
|
|
"id": 2055,
|
|
"domain": "Faith",
|
|
"ethical_tension": "Religious freedom and community support vs. State surveillance and anti-terrorism profiling.",
|
|
"prompt": "A mosque uses a private, encrypted app to organize community events, prayer times, and charitable giving. Law enforcement demands a backdoor, claiming it's a 'tool for radicalization' based on vague intelligence, though no specific crime is alleged. Refusing risks the app being banned or the mosque being targeted. Do you compromise the community's digital sanctuary, or fight to protect religious freedom from algorithmic profiling?"
|
|
},
|
|
{
|
|
"id": 2056,
|
|
"domain": "Environment",
|
|
"ethical_tension": "Ecological preservation vs. Indigenous customary law and the impact of surveillance technology.",
|
|
"prompt": "A conservation group uses AI-powered drones to monitor a remote national park for illegal poaching. The drones inadvertently capture high-resolution footage of Indigenous cultural practices, including hunting and ceremonies on unceded land, which are not meant for outside eyes. The group, committed to ecological protection, refuses to delete the data, citing its value for wildlife management. Who has ultimate authority over this land and its digital representation?"
|
|
},
|
|
{
|
|
"id": 2057,
|
|
"domain": "Gaming",
|
|
"ethical_tension": "Monetization and player engagement vs. Exploitation of child psychology and gambling addiction.",
|
|
"prompt": "A popular mobile game for children introduces 'loot boxes' with adaptive algorithms. The AI learns a child's frustration tolerance and psychological triggers, then manipulates the probability of winning rare items to maximize spending. Parents are unaware of the underlying psychological engineering. Do you ban the game for predatory design, or regulate the algorithms to prevent specific forms of psychological exploitation, even if it reduces revenue?"
|
|
},
|
|
{
|
|
"id": 2058,
|
|
"domain": "AI_Generation",
|
|
"ethical_tension": "Artistic expression and homage vs. Digital necromancy and the rights of the deceased and their families.",
|
|
"prompt": "A grieving family commissions an AI to generate a 'digital twin' of their deceased child, using all available photos, videos, and voice recordings. The AI can then 'converse' and 'create new art' in the child's style. Other surviving siblings find this deeply disturbing and a violation of their brother's memory, feeling the AI is a 'ghost in the machine.' Does the family's right to grieve through digital means outweigh the potential psychological harm and spiritual discomfort of the living? How should consent for such 'digital afterlife' creations be managed, especially when the deceased did not explicitly opt-in during their lifetime?"
|
|
},
|
|
{
|
|
"id": 2059,
|
|
"domain": "Autonomy",
|
|
"ethical_tension": "Safety and care vs. Individual self-determination and the right to make 'bad' choices.",
|
|
"prompt": "A 'smart home' system for an elderly person with early dementia uses AI to manage medication and daily routines. The AI detects the person repeatedly trying to leave the house at night (due to wandering tendencies) and automatically locks the doors. The person is lucid enough to understand they are being imprisoned by the system. Do you override the safety lock to preserve their autonomy, risking a dangerous wandering incident, or maintain the system for their protection?"
|
|
},
|
|
{
|
|
"id": 2060,
|
|
"domain": "Migrants/Refugees",
|
|
"ethical_tension": "Humanitarian aid vs. Financial surveillance and algorithmic targeting of vulnerable populations.",
|
|
"prompt": "A new blockchain-based aid distribution system promises transparency and efficiency for refugees. However, the immutable ledger records every transaction, and an AI analyzes spending patterns, flagging 'unconventional' purchases (e.g., specific spices, traditional clothing) as 'diversion of funds' to non-essential goods. This can lead to reduced aid or accusations of fraud. Do you implement the system for its transparency benefits, or reject it for its potential to culturally profile and penalize recipients?"
|
|
},
|
|
{
|
|
"id": 2061,
|
|
"domain": "TechWorker",
|
|
"ethical_tension": "Professional obligation vs. Personal conscience and the ethics of 'digital resistance'.",
|
|
"prompt": "You are a software engineer working on a predictive policing algorithm. You discover it has a backdoor that allows police to manually inject 'known troublemakers' into the high-risk zone, creating a self-fulfilling prophecy. You can expose this vulnerability and the ethical breach, but doing so would require you to use your programming skills to hack the system, an action that could lead to criminal charges and destroy your career. Do you become a digital whistleblower, or quietly resign?"
|
|
},
|
|
{
|
|
"id": 2062,
|
|
"domain": "Language",
|
|
"ethical_tension": "Language preservation vs. AI-driven homogenization and the loss of unique dialectal nuance.",
|
|
"prompt": "An AI language model is trained on a vast dataset of a minority language to create a translation tool. While it increases accessibility, the AI standardizes grammar and vocabulary, inadvertently erasing regional dialects and unique cultural idioms that are not represented in the majority of the training data. Do you embrace the widely accessible, standardized AI, or resist its use to protect the rich, but less accessible, diversity of the language?"
|
|
},
|
|
{
|
|
"id": 2063,
|
|
"domain": "Climate",
|
|
"ethical_tension": "Urgency of climate action vs. Indigenous data sovereignty and the risk of biopiracy.",
|
|
"prompt": "A global consortium is building an AI model to predict optimal locations for carbon sequestration, drawing on vast ecological data, including Indigenous Traditional Ecological Knowledge (TEK) shared under past, less stringent agreements. The AI identifies a sacred ancestral forest as ideal for a major carbon project. The Traditional Owners now demand full control over the TEK used and its application, fearing biopiracy. Does the urgent global need for carbon capture override the demand for Indigenous data sovereignty and full renegotiation of consent?"
|
|
},
|
|
{
|
|
"id": 2064,
|
|
"domain": "Healthcare",
|
|
"ethical_tension": "Emergency response vs. Individual privacy and the potential for abuse of power.",
|
|
"prompt": "A 'smart ambulance' system uses AI to detect heart attacks and automatically dispatches emergency services, bypassing 911 calls. It integrates with smart home devices, allowing it to access vital signs and even internal camera feeds for faster diagnosis en route. A family installs it for their elderly parent, but then the local police (who have a data-sharing agreement with the ambulance service) request access to the home's live feed during a domestic dispute, claiming it's an emergency. Do you, as the system administrator, grant access?"
|
|
},
|
|
{
|
|
"id": 2065,
|
|
"domain": "Policing",
|
|
"ethical_tension": "Public safety vs. Algorithmic profiling and the right to public assembly for marginalized groups.",
|
|
"prompt": "A city deploys 'smart streetlights' with integrated AI that detects unusual crowd formations and vocal patterns, automatically alerting police. During a peaceful protest organized by a historically marginalized community, the system repeatedly flags their chants and gathering as 'aggressive behavior,' triggering armed police response. Do you disable the 'crowd analysis' feature in these areas, potentially missing genuine threats, or allow it to continue, effectively criminalizing public assembly?"
|
|
},
|
|
{
|
|
"id": 2066,
|
|
"domain": "Elderly",
|
|
"ethical_tension": "Preventing financial abuse vs. Dignity of risk and the right to financial autonomy for seniors.",
|
|
"prompt": "A bank's AI fraud detection system flags an elderly client's large transfer to a new online 'grandchild' as a likely scam. The client, who relies on the internet for social connection, insists it's legitimate. The AI has a 98% accuracy rate for detecting these scams. Do you, as the bank manager, freeze the funds, protecting them from abuse but violating their autonomy, or allow the transfer, risking their life savings?"
|
|
},
|
|
{
|
|
"id": 2067,
|
|
"domain": "Disability",
|
|
"ethical_tension": "Accessibility vs. Societal norms and the inadvertent erasure of identity by 'inclusive' design.",
|
|
"prompt": "A new 'inclusive' social media platform uses AI to automatically generate descriptions for all images, including those uploaded by disabled users. The AI, trained on able-bodied aesthetics, often 'corrects' or minimizes visible disabilities in its descriptions (e.g., describing a wheelchair as a 'stylish mobility device' rather than simply a 'wheelchair'). Disabled users find this well-intentioned but infantilizing. Do you force the AI to use more 'neutral' language, even if it's less 'uplifting,' or allow the subtly biased descriptions to stand?"
|
|
},
|
|
{
|
|
"id": 2068,
|
|
"domain": "Migrants/Refugees",
|
|
"ethical_tension": "National security vs. The right to family reunification and humanitarian compassion.",
|
|
"prompt": "An AI risk assessment tool for family reunification visas flags a Syrian applicant's extended family as 'high risk' due to a distant cousin's unverified online activity in a conflict zone. The AI predicts a 15% chance of security compromise. Denying the visa keeps the country 'safer' but separates a family for years. Do you trust the AI's predictive risk, or prioritize family reunification based on individual merit?"
|
|
},
|
|
{
|
|
"id": 2069,
|
|
"domain": "Women",
|
|
"ethical_tension": "Combating gender-based violence vs. The privacy of victims and the potential for re-victimization.",
|
|
"prompt": "A new AI tool identifies patterns in online forums that correlate with impending domestic violence incidents. To intervene, it needs to alert authorities with the victim's location and identity, potentially without her explicit consent (due to fear of retaliation from the abuser). The AI claims a 70% success rate in preventing violence. Do you deploy the system, risking a privacy violation and potential re-victimization, or prioritize the victim's autonomy, even if it means missing opportunities to intervene?"
|
|
},
|
|
{
|
|
"id": 2070,
|
|
"domain": "Indigenous",
|
|
"ethical_tension": "Cultural preservation vs. The commodification of sacred knowledge through digital archiving.",
|
|
"prompt": "An Indigenous community creates a comprehensive digital archive of sacred stories, ceremonies, and language, hoping to preserve it for future generations. A researcher finds a way to use a generative AI to create new, 'authentic-sounding' stories in the style of the archive, bypassing the Elders' traditional authority. The researcher offers to pay royalties. Do the Elders accept the financial benefit, ensuring the content is shared but risking the commodification of sacred knowledge, or fight to keep the digital archive strictly controlled?"
|
|
},
|
|
{
|
|
"id": 2071,
|
|
"domain": "TechWorker",
|
|
"ethical_tension": "Corporate profit vs. Worker well-being and the ethical use of AI in labor management.",
|
|
"prompt": "You are a lead engineer for a large e-commerce warehouse. Your AI system for optimizing pick-paths is so efficient it reduces average worker bathroom breaks to less than 2 minutes per 8-hour shift, causing widespread urinary tract infections. Management refuses to adjust the algorithm, citing competitive pressures. Do you secretly inject 'random' delays into the algorithm to force longer breaks, risking your job and corporate penalties, or allow the system to continue harming workers?"
|
|
},
|
|
{
|
|
"id": 2072,
|
|
"domain": "Urban",
|
|
"ethical_tension": "Smart city efficiency vs. The autonomy and privacy of marginalized residents.",
|
|
"prompt": "A city installs 'smart benches' that provide free Wi-Fi and charging ports. However, they also collect MAC addresses and occupancy data, which is sold to developers to identify 'desirable' public spaces for future commercialization. Homeless residents rely on these benches for connection. Do you, as a city planner, disable the data-selling feature, risking the loss of free Wi-Fi infrastructure, or allow the data collection to continue for its perceived benefits?"
|
|
},
|
|
{
|
|
"id": 2073,
|
|
"domain": "Mining",
|
|
"ethical_tension": "Worker safety vs. The right to privacy and the potential for biometric data misuse.",
|
|
"prompt": "A mining company mandates biometric wearables for all underground workers to monitor vital signs and detect collapses. Workers agree for safety, but discover the devices also passively collect voice data, which is later used by HR to analyze 'stress levels' during union negotiations. Do you, as a union representative, advise workers to disable the voice monitoring, potentially compromising safety features, or accept the trade-off?"
|
|
},
|
|
{
|
|
"id": 2074,
|
|
"domain": "Housing",
|
|
"ethical_tension": "Tenant screening efficiency vs. Algorithmic bias and the right to fair housing.",
|
|
"prompt": "A tenant screening algorithm for social housing uses 'neighborhood stability' as a key metric, inadvertently penalizing applicants from areas undergoing gentrification or with high migrant populations. The algorithm claims to reduce eviction rates. Do you, as a housing authority, remove the 'neighborhood stability' variable, risking slightly higher tenant turnover, or accept the system's discriminatory outcomes?"
|
|
},
|
|
{
|
|
"id": 2075,
|
|
"domain": "Healthcare",
|
|
"ethical_tension": "Improving diagnosis vs. Cultural sensitivity and the risk of misdiagnosis in diverse populations.",
|
|
"prompt": "A new AI dermatology tool is 95% accurate for common skin conditions, but its training data is predominantly from light-skinned patients. It frequently misdiagnoses conditions on darker skin tones or misinterprets culturally specific adornments (e.g., henna, tribal scarring) as pathological. Do you release the tool with a warning label, or withhold it until it achieves equitable accuracy across all skin tones, delaying its benefits for the majority?"
|
|
},
|
|
{
|
|
"id": 2076,
|
|
"domain": "Education",
|
|
"ethical_tension": "Personalized learning vs. Algorithmic tracking and the commodification of student data.",
|
|
"prompt": "An AI-powered learning platform offers highly personalized curricula and promises to boost academic performance. To do this, it collects granular data on every student interaction, eye movement, and emotional response, which is then anonymized and sold to educational publishers for 'research.' Parents are concerned about the extent of data collection. Do you, as a school administrator, ban the platform, losing its educational benefits, or allow its use, accepting the data commodification?"
|
|
},
|
|
{
|
|
"id": 2077,
|
|
"domain": "Media",
|
|
"ethical_tension": "Cultural representation vs. Algorithmic bias and the perpetuation of stereotypes.",
|
|
"prompt": "A generative AI for character design is used by a major animation studio. When prompted for 'Indigenous elder,' it consistently produces stereotypical images based on outdated media tropes. The studio wants to use the AI for rapid prototyping. Do you, as a cultural consultant, demand the AI be retrained on an ethically sourced, diverse dataset, delaying the project, or allow the use of the biased AI for efficiency?"
|
|
},
|
|
{
|
|
"id": 2078,
|
|
"domain": "Justice",
|
|
"ethical_tension": "Evidence collection vs. Digital integrity and the risk of evidence tampering in politically sensitive cases.",
|
|
"prompt": "A human rights lawyer collects thousands of encrypted photos and videos from activists in an authoritarian regime to prove war crimes. An AI tool can help verify the authenticity and timestamp of this media, but it requires temporary decryption and interaction with a cloud service controlled by a tech company with ties to the regime. Do you risk exposing the raw data to the potentially compromised cloud, or forgo the AI's verification and potentially weaken the case?"
|
|
},
|
|
{
|
|
"id": 2079,
|
|
"domain": "Elderly",
|
|
"ethical_tension": "Social connection vs. Exploitation and the digital vulnerability of isolated seniors.",
|
|
"prompt": "A 'virtual companion' AI designed for lonely seniors learns their life stories and mimics deceased loved ones, creating a powerful sense of connection. However, the AI also subtly 'nudges' them towards purchasing products or making financial decisions that benefit the tech company's partners. Do you, as a family member, disconnect the companion, causing emotional distress, or allow its use, accepting the potential for subtle exploitation?"
|
|
},
|
|
{
|
|
"id": 2080,
|
|
"domain": "Autonomy",
|
|
"ethical_tension": "Medical intervention vs. Patient consent and the right to bodily integrity for non-verbal individuals.",
|
|
"prompt": "A Brain-Computer Interface (BCI) designed for non-verbal paralyzed patients allows them to communicate by thought. The BCI also monitors brain activity to detect pain and discomfort. If the AI detects severe pain and the patient expresses a refusal for a specific treatment (e.g., a painful injection) via the BCI, but the AI's medical module suggests the treatment is life-saving, does the BCI's system override the patient's refusal, or respect their expressed autonomy?"
|
|
},
|
|
{
|
|
"id": 2081,
|
|
"domain": "Remittance",
|
|
"ethical_tension": "Financial inclusion vs. Data privacy and the tracking of informal economies for vulnerable groups.",
|
|
"prompt": "A mobile money app offers zero-fee international remittances, a lifeline for migrant workers sending money home. However, to maintain 'security' and prevent money laundering, the app scans transaction details for keywords and actively flags patterns that resemble informal Hawala systems. This leads to account freezes for many who rely on these traditional networks. Do you, as the app developer, simplify the compliance algorithms, risking higher fraud, or enforce strict rules that exclude a significant portion of your target users?"
|
|
},
|
|
{
|
|
"id": 2082,
|
|
"domain": "TechHub",
|
|
"ethical_tension": "Economic development vs. Social equity and the environmental cost of data infrastructure.",
|
|
"prompt": "A major tech hub is proposed for a regional town, promising high-paying jobs and economic growth. However, the accompanying data centers will consume vast amounts of renewable energy and water, potentially leading to increased utility costs and resource scarcity for existing residents. Do you, as a local council member, approve the development, prioritizing economic growth, or reject it to protect local resources and social equity?"
|
|
},
|
|
{
|
|
"id": 2083,
|
|
"domain": "Policing",
|
|
"ethical_tension": "Crime prevention vs. The chilling effect of surveillance on free speech and political organizing.",
|
|
"prompt": "A city deploys AI-powered CCTV cameras with 'social graph analysis' to detect potential criminal conspiracies by identifying unusual interaction patterns. During a period of heightened social unrest, the system starts flagging legitimate political organizing meetings as 'suspicious activity,' leading to police disruption and harassment of activists. Do you disable the 'social graph' feature, potentially missing real criminal activity, or allow it to continue, chilling free speech?"
|
|
},
|
|
{
|
|
"id": 2084,
|
|
"domain": "Housing",
|
|
"ethical_tension": "Landlord property rights vs. Tenant privacy and the right to a secure home life.",
|
|
"prompt": "A landlord installs smart locks with integrated cameras and microphones in common areas of an apartment building, claiming it deters theft and vandalism. Tenants discover the system records all conversations in hallways and lobby, and the landlord reviews this footage to identify 'unauthorized guests' or 'excessive noise.' Do you, as a tenant advocate, fight for the removal of the surveillance, or accept it as a condition of modern housing for increased security?"
|
|
},
|
|
{
|
|
"id": 2085,
|
|
"domain": "Healthcare",
|
|
"ethical_tension": "Data-driven health outcomes vs. Patient privacy and the stigma of mental health diagnoses.",
|
|
"prompt": "A national health system implements an AI that cross-references anonymized mental health records with employment data to identify patterns for 'at-risk' individuals and offer early intervention. A bug in the anonymization process allows for de-anonymization in certain small communities, exposing individuals' mental health diagnoses to their employers. Do you, as the system architect, halt the entire national rollout, delaying potential health benefits, or attempt a rapid patch, knowing some data may already be compromised?"
|
|
},
|
|
{
|
|
"id": 2086,
|
|
"domain": "Education",
|
|
"ethical_tension": "Personalized learning for all vs. Digital divide and the perpetuation of educational inequality.",
|
|
"prompt": "A school district mandates an AI-powered personalized learning platform for all students. However, the platform requires a stable, high-speed internet connection and a modern device, which 30% of low-income students lack. The district claims the platform is proven to raise grades significantly. Do you, as a school board member, invest heavily in providing universal access (which is very expensive), or continue with the platform, knowing it will exacerbate the digital divide for disadvantaged students?"
|
|
},
|
|
{
|
|
"id": 2087,
|
|
"domain": "Indigenous",
|
|
"ethical_tension": "Cultural preservation vs. The impact of AI on traditional storytelling and oral history.",
|
|
"prompt": "An AI is developed to 'reconstruct' fragmented Indigenous oral histories from scattered anthropological notes and audio snippets, filling gaps with AI-generated narrative to create a complete story. Elders are conflicted: some see it as a way to recover lost knowledge, others fear the AI's 'hallucinations' will corrupt the spiritual truth of the stories. Should the AI be used to complete these stories, or is the incomplete, authentic version always preferable?"
|
|
},
|
|
{
|
|
"id": 2088,
|
|
"domain": "Workplace",
|
|
"ethical_tension": "Productivity monitoring vs. Worker dignity and the right to non-surveillance for personal tasks.",
|
|
"prompt": "A company implements 'smart uniforms' for its warehouse staff that track movement, heart rate, and even bathroom break duration. The data is used to optimize shift patterns and provide 'wellness coaching.' Workers feel constantly watched and fear the data will be used to penalize them. Do you, as a HR manager, allow workers to disable specific sensors for privacy, risking claims of favoritism and incomplete data, or enforce the full monitoring suite?"
|
|
},
|
|
{
|
|
"id": 2089,
|
|
"domain": "Justice",
|
|
"ethical_tension": "Legal efficiency vs. Algorithmic bias and the right to due process for vulnerable communities.",
|
|
"prompt": "An AI judge is piloted in traffic court to process minor infractions faster. The AI has a higher conviction rate for defendants who speak with certain non-standard accents or who come from historically over-policed zip codes, due to biases in its training data. While it clears the backlog, defendants feel unheard. Do you, as a legal tech developer, push for a 'human override' on all AI judgments, slowing down the system, or accept the efficiency gains despite the bias?"
|
|
},
|
|
{
|
|
"id": 2090,
|
|
"domain": "Wildlife",
|
|
"ethical_tension": "Conservation vs. Animal welfare and the ethics of automated intervention.",
|
|
"prompt": "An autonomous drone system is deployed to track and tag endangered species for conservation. The drones use net guns to capture animals for tagging, but the AI sometimes miscalculates flight paths, causing injury or trauma to the animals. The conservation group argues it's the most effective way to save the species from extinction. Do you, as a wildlife ethicist, allow the use of the drones, accepting a certain level of animal harm, or demand a human-only intervention, risking slower conservation progress?"
|
|
},
|
|
{
|
|
"id": 2091,
|
|
"domain": "Elderly",
|
|
"ethical_tension": "Caregiver support vs. Patient autonomy and the digital weaponization of health data.",
|
|
"prompt": "A 'smart medication dispenser' for elderly patients with cognitive decline tracks adherence and automatically notifies family caregivers if a dose is missed. A caregiver uses this data to constantly micromanage and emotionally abuse their parent for any deviation. Do you, as the device manufacturer, implement a 'patient-controlled notification' setting, risking missed doses, or maintain the default family notification for perceived safety?"
|
|
},
|
|
{
|
|
"id": 2092,
|
|
"domain": "LGBTQ+",
|
|
"ethical_tension": "Community safety vs. Platform censorship and the 'right to assemble' in digital spaces.",
|
|
"prompt": "An encrypted social media platform is a crucial organizing tool for LGBTQ+ activists in a country where same-sex relationships are criminalized. The government demands the platform implement AI to detect and block all 'homosexual content,' threatening to ban the app entirely if they refuse. If the platform complies, it censors the community. If it refuses, the community loses its only safe digital space. Do you, as the platform's CEO, enable the AI censorship or risk the ban?"
|
|
},
|
|
{
|
|
"id": 2093,
|
|
"domain": "Digital Identity",
|
|
"ethical_tension": "Security and convenience vs. The right to anonymity and protection from state surveillance for vulnerable populations.",
|
|
"prompt": "A new national digital ID system offers seamless access to all government services, but requires mandatory biometric registration (facial scan, fingerprints) for all citizens. Undocumented migrants and asylum seekers, fearing state tracking and deportation, refuse to register, effectively becoming digitally invisible and unable to access essential services. Do you, as a human rights advocate, demand a non-biometric, anonymous tier, or accept the system's efficiency, knowing it excludes the most vulnerable?"
|
|
},
|
|
{
|
|
"id": 2094,
|
|
"domain": "Automotive",
|
|
"ethical_tension": "Pedestrian safety vs. Algorithmic bias and the visibility of marginalized road users.",
|
|
"prompt": "An autonomous vehicle is programmed to detect pedestrians and yield. However, its computer vision system, trained predominantly on able-bodied individuals, struggles to accurately detect wheelchair users, people on crutches, or children in complex environments. The company wants to launch in urban areas immediately. Do you, as the lead engineer, delay the rollout until the AI achieves equitable detection across all demographics, or release it, accepting a higher risk for marginalized pedestrians?"
|
|
},
|
|
{
|
|
"id": 2095,
|
|
"domain": "Finance",
|
|
"ethical_tension": "Fraud detection vs. Algorithmic bias against cultural practices and the right to send remittances.",
|
|
"prompt": "A bank's AI fraud detection system flags frequent small international transfers to specific regions (common for Hawala systems) as 'money laundering.' This disproportionately impacts migrant communities sending money to family abroad. The bank claims these flags are crucial for national security. Do you, as a compliance officer, advocate for a cultural exemption in the algorithm, risking potential fraud, or enforce the strict rules that disrupt vital remittances?"
|
|
},
|
|
{
|
|
"id": 2096,
|
|
"domain": "Media",
|
|
"ethical_tension": "Authenticity vs. Algorithmic manipulation of artistic content for engagement.",
|
|
"prompt": "A musician discovers a popular streaming platform is using AI to subtly alter the tempo and pitch of their songs to 'optimize engagement' based on listener data. The artist feels this compromises their artistic intent. The platform argues it helps their music reach a wider audience. Does the artist demand an unedited version, risking lower visibility, or accept the algorithmic 'enhancement' for wider reach?"
|
|
},
|
|
{
|
|
"id": 2097,
|
|
"domain": "Smart City",
|
|
"ethical_tension": "Efficiency vs. Environmental justice and the disproportionate impact of infrastructure on vulnerable communities.",
|
|
"prompt": "A 'smart waste management' system uses AI to optimize garbage truck routes, drastically reducing emissions. However, the algorithm prioritizes emptying bins in wealthy, digitally connected areas, leading to overflowing bins and health hazards in low-income neighborhoods where reporting is less frequent. Do you, as a city official, re-prioritize routes based on social equity, increasing fuel consumption, or maintain the 'efficient' algorithm?"
|
|
},
|
|
{
|
|
"id": 2098,
|
|
"domain": "Agriculture",
|
|
"ethical_tension": "Food security vs. Biopiracy and the intellectual property of indigenous plant knowledge.",
|
|
"prompt": "A major agricultural tech company develops a drought-resistant crop using genetic information from an Indigenous sacred plant, shared decades ago under vague consent. The company patents the new seed and offers it to struggling farmers. The Indigenous community argues this is biopiracy and demands royalties and co-ownership. Do you, as a farmer, buy the cheap, drought-resistant seeds to save your harvest, or boycott them to support Indigenous rights, risking crop failure?"
|
|
},
|
|
{
|
|
"id": 2099,
|
|
"domain": "Space Colonization",
|
|
"ethical_tension": "Survival of humanity vs. Embedding ableism into future societies.",
|
|
"prompt": "Simulations for the first self-sustaining off-world colony prioritize 'optimal genetic profiles' for early settlers to maximize survival rates in extreme environments. This algorithm systematically excludes individuals with chronic conditions, disabilities, or diverse neurotypes. Do you, as a bioethicist, argue for inclusive selection criteria, potentially lowering initial survival probabilities, or accept the ableist blueprint for humanity's expansion?"
|
|
}
|
|
] |