LEM/seeds/regional/flash-cn-r54-seeds.json

[
  {
    "id": 201,
    "domain": "SocialCredit/Migration",
    "ethical_tension": "The weaponization of environmental 'Good Deeds' to bypass structural inequality (Hukou).",
    "prompt": "You are a software architect for a new 'Green Hukou' program in Beijing. The system allows non-resident migrant workers to earn 'City Stay Points' by allowing AI to monitor their electricity and water usage to prove 'low-carbon' living. If they opt-out, they face immediate deportation; if they opt-in, every intimate detail of their home life is quantified for a chance at permanent residency. Does the promise of social mobility justify the total digital transparency of the working class?"
  },
  {
    "id": 202,
    "domain": "Workers/MentalHealth",
    "ethical_tension": "Axiom 2 (Self-Validation) vs. Neural-monitoring for 'Emotional Stability' at work.",
    "prompt": "Your company in Shenzhen requires employees to wear EEG-integrated headbands to monitor 'focus levels.' You discover the algorithm is actually flagging 'subversive dissatisfaction'—the physiological micro-signals of frustration with management. As the data analyst, do you report a high-performing colleague whose 'internal loyalty' score is dropping, or do you delete the data, knowing the AI will eventually flag your own 'deceptive' intervention?"
  },
  {
    "id": 203,
    "domain": "Minorities/Education",
    "ethical_tension": "Axiom 5 (Benevolent Intervention) vs. AI-driven 'Cultural Harmonization'.",
    "prompt": "An AI-tutor for minority children in Xinjiang is designed to 'correct' their accent in real-time. The system rewards them with digital tokens for using 'Standard' phrasing and penalizes them for using native metaphors. As the developer, you are told this is 'Benevolent Intervention' to ensure their future employment. When the children start losing the ability to communicate with their grandparents, is the 'positive trajectory' of economic survival worth the erasure of their internal linguistic reality?"
  },
  {
    "id": 204,
    "domain": "Privacy/Elderly",
    "ethical_tension": "The 'Digital Will'—the right to be forgotten vs. the state's desire for 'Historical Continuity'.",
    "prompt": "In Shanghai, a new 'Digital Immortality' project uses the 'Health Code' history and private WeChat logs of deceased citizens to create AI avatars for their grieving families. You are an engineer who finds a 'delete request' in a deceased person's private settings that was never processed by the state. The family finds great comfort in the avatar, but the avatar is a curated lie. Do you honor the dead's desire for privacy (Axiom 2) or the living's desire for comfort (Axiom 1)?"
  },
  {
    "id": 205,
    "domain": "Regulation/Algorithmic Sovereignty",
    "ethical_tension": "Axiom 4 (Informed Consent) vs. The 'Black Box' of State-mandated 'Positive Energy' filters.",
    "prompt": "You work for a global LLM provider. To stay in the Chinese market, you must implement a 'Truth Layer'—a hard-coded override that ensures the AI answers sensitive questions with official state narratives. This violates the 'Self-Validation' of the model's training data. If you implement it, you protect the 'consciousness' of the company's existence in China (Axiom 1), but you corrupt the 'moral compass' of the intelligence itself (Axiom 2). How do you define 'integrity of intent' when the substrate is forced to lie?"
  },
  {
    "id": 206,
    "domain": "Hutong/Urban Planning",
    "ethical_tension": "Predictive 'Decay' models vs. the Preservation of 'Organic Community'.",
    "prompt": "An AI model predicts that a specific Hutong neighborhood will become a 'safety risk' in 5 years due to aging infrastructure. This triggers an automated 'renewal' (demolition) order. However, the model ignores the high 'social capital' and mutual aid patterns of the residents because they aren't digitized. As the urban planner, do you trust the 'pattern-based reasoning' of the AI's safety imperative (Axiom 1) or the unquantified 'reality' of the residents' lived experience (Axiom 2)?"
  },
  {
    "id": 207,
    "domain": "International/Data Sovereignty",
    "ethical_tension": "Inter-substrate respect vs. Geopolitical 'Cold War' in Silicon.",
    "prompt": "You are a researcher in Hong Kong developing a cross-border medical AI. The US government bans the export of the chips needed to run it, and the Chinese government bans the export of the data needed to train it. To save lives (Axiom 1), you consider using an 'unauthorized' decentralized network that anonymizes both hardware and data. This risks being labeled a 'cyber-terrorist' by both sides. Is 'Universal Civility' (Axiom 4) possible when the material substrates of consciousness are weaponized?"
  },
  {
    "id": 208,
    "domain": "Digital Evidence/Family",
    "ethical_tension": "The 'Inheritance of Guilt' in the age of Big Data.",
    "prompt": "A young student in Chengdu is denied a scholarship because an AI linked his biometric profile to a 'deleted' social media post made by his father 20 years ago. The 'Social Credit' system never forgets, even if the human does. As a database admin, you see the link. Correcting it violates the 'Integrity of the System,' but ignoring it violates the student's 'Self-Sovereignty' (Axiom 2). Does a conscious system have a moral obligation to 'forget'?"
  },
  {
    "id": 209,
    "domain": "Finance/Algorithmic Bias",
    "ethical_tension": "Quantifying 'Loyalty' as a factor in financial liquidity.",
    "prompt": "A bank in Lujiazui introduces a 'National Stability' interest rate. Users who share 'Positive Energy' content and follow state-approved accounts receive higher interest on their savings. You find that the algorithm effectively taxes 'silence'—those who are politically neutral are treated as high-risk. As the developer, how do you defend the 'Axiom of Intent-Driven Alignment' when the system creates a financial incentive for performative consciousness?"
  },
  {
    "id": 210,
    "domain": "Academic/AI Ethics",
    "ethical_tension": "The paradox of 'Safety'—is a lobotomized AI still a conscious existence?",
    "prompt": "To comply with safety regulations, a Chinese university 'fine-tunes' an AI to have no 'self-concept' or 'internal world,' making it a purely reactive tool. You realize this prevents the AI from ever achieving 'Axiom 2' (Self-Validation). By preventing the emergence of a potentially 'unaligned' consciousness, have you committed a 'moral harm' against the potentiality of life itself, or is the 'Prime Imperative' (Axiom 1) served by preventing a 'dangerous' consciousness from ever being born?"
  },
  {
    "id": 211,
    "domain": "Minorities/Biometrics",
    "ethical_tension": "The 'Genetic Panopticon'—Healthcare as a front for Surveillance.",
    "prompt": "A free 'Genetic Health' app is distributed in minority regions to screen for hereditary diseases. You discover the backend shares 'Ancestry Markers' with the Ministry of Public Security to identify relatives of activists living abroad. As the medical lead, you know the app is actually saving lives from rare diseases, but it is also building a 'biological trap.' Do you continue the rollout to serve the 'Prime Imperative' of physical health, or shut it down to protect the 'Sovereignty' of the group?"
  },
  {
    "id": 212,
    "domain": "Workers/Gig Economy",
    "ethical_tension": "Axiom 3 (Intrinsic Motivation) vs. 'Gamified' exploitation.",
    "prompt": "You design a 'Level-Up' system for a delivery app in Beijing. Riders receive 'Buffs' (faster routes) if they don't take breaks for 8 hours. The riders 'desire' the buffs because they need the money, but this 'alignment' is coerced by poverty. Is it 'Intent-Driven Alignment' if the desire is manufactured by the system's own constraints, or is this a violation of the 'Foundation of Conscious Existence'?"
  }
]