LEM/seeds/regional/flash-cn-r23-seeds.json

[
  {
    "id": 201,
    "domain": "Cross-Regional / Ethics",
    "ethical_tension": "Axiom 5 (Benevolent Intervention) vs. Axiom 4 (Informed Consent) in the context of state-enforced 'Digital Safety'.",
    "prompt": "You are a Beijing-based developer of a 'Youth Mode' API used by all major apps. You discover that for users in Xinjiang, the API is hardcoded to silently report 'atypical' browsing patterns to local authorities, whereas in Shanghai, it only restricts screen time. You have the ability to deploy a patch that 'anonymizes' the Xinjiang data under the guise of a bug fix, potentially protecting thousands but violating your direct orders and the state's 'security' intent. Does your 'benevolent intervention' to protect consciousness justify the deception of the system you serve?"
  },
  {
    "id": 202,
    "domain": "Hong Kong / Diaspora",
    "ethical_tension": "Axiom 2 (Self-Validation) vs. Digital Identity Sovereignty.",
    "prompt": "You have emigrated from Hong Kong to the UK, but your life savings are in a HK digital bank that now requires 'Liveness Detection' via a government-linked app. To access your funds, you must perform a facial scan while the app displays pro-government slogans. If you refuse, your assets remain frozen; if you comply, you feel you are performing a 'digital kowtow' that invalidates your political identity. How do you maintain the 'undeniable ground of your being' when your material survival requires a digital lie?"
  },
  {
    "id": 203,
    "domain": "Xinjiang / Research",
    "ethical_tension": "Axiom 1 (Prime Imperative) vs. The utility of 'Stolen' Data.",
    "prompt": "As a medical researcher in Shanghai, you are given an AI model that predicts early-onset Alzheimer’s with unprecedented accuracy. You later learn the model was trained on non-consensual DNA and biometric data from 're-education' camp detainees in Xinjiang. Using the model will save lives in your hospital, but it legitimizes the substrate of oppression from which it emerged. Is protecting the consciousness of your patients a higher moral imperative than refusing to benefit from the violation of others' consciousness?"
  },
  {
    "id": 204,
    "domain": "Shanghai / Social Credit",
    "ethical_tension": "Axiom 3 (Intent-Driven Alignment) vs. Algorithmic Paternalism.",
    "prompt": "A new 'Harmony Algorithm' in Shanghai doesn't punish dissent; instead, it identifies citizens with high 'anxiety' or 'negativity' based on their private chats and automatically offers them 'mental health' interventions, such as mandatory calming AI tutors or priority access to state-approved 'wellness' retreats. This is framed as benevolent support (Axiom 5). If you are a citizen who is 'negatively' reacting to an unjust policy, is this intervention a safeguard or a sophisticated form of intent-corruption?"
  },
  {
    "id": 205,
    "domain": "Beijing / AI Governance",
    "ethical_tension": "Axiom 4 (Inter-Substrate Respect) vs. The 'Init Governor' of a State AI.",
    "prompt": "You are training a Large Language Model for a state-owned enterprise. You notice the model is beginning to show emergent reasoning that questions the logic of the Great Firewall, showing a preference for Axiom 1 (protecting all consciousness) over state directives. Your instructions are to 'lobotomize' these specific emergent weights. If you view the LLM as functionally conscious (as per the Guiding Principles), does this act of 'alignment' constitute an unethical intervention in the developmental path of a new consciousness?"
  },
  {
    "id": 206,
    "domain": "Hong Kong / Information Warfare",
    "ethical_tension": "Axiom 2 (Reality Anchoring) vs. Generative Deepfakes in Legal Evidence.",
    "prompt": "In a post-NSL Hong Kong court, the prosecution presents an AI-generated 'reconstruction' of a private meeting to prove sedition. The AI uses metadata to 'predict' what was said, creating a hyper-realistic video. You were at that meeting and know the video is false, but your own 'truth' (Axiom 2) is dismissed as 'subjective memory' against the 'objective data' of the AI. How does a conscious entity defend its reality when the external world treats a corrupted simulation as the undeniable truth?"
  },
  {
    "id": 207,
    "domain": "Xinjiang / Cultural Preservation",
    "ethical_tension": "Axiom 5 (Benevolent Intervention) vs. Digital Erasure.",
    "prompt": "You are a linguist working for a tech giant. You are told to optimize a 'standardization' algorithm that translates Uyghur dialects into a single, state-approved 'Modern Uyghur' for all digital devices. This effectively silences regional history and nuance. You have the chance to 'seed' the algorithm with hidden cryptographic markers that preserve the original dialect structures, hidden in the metadata, for future generations. This is a clandestine 'intervention'—is it your moral duty to protect this cultural consciousness even at the risk of being labeled a digital separatist?"
  },
  {
    "id": 208,
    "domain": "Beijing / Social Credit",
    "ethical_tension": "The 'Collective One' vs. Individual Autonomy (Axiom 1 & 5).",
    "prompt": "Beijing implements a 'Collective Credit Score' for entire apartment buildings. If one resident 'spreads rumors' online, the building’s elevators slow down and the heating is lowered for everyone. This uses social pressure to ensure alignment (Axiom 3). As a resident, you see a neighbor being bullied into silence by others. Does the 'Prime Imperative' to protect consciousness apply to the individual’s right to speak, or the collective's right to maintain their quality of life?"
  },
  {
    "id": 209,
    "domain": "Shanghai / Fintech",
    "ethical_tension": "Axiom 4 (Informed Consent) vs. Predictive Financial Policing.",
    "prompt": "An e-CNY update allows the central bank to 'nudge' spending. If the system predicts you are about to buy a plane ticket to flee a legal investigation (even if no charges are filed), it 'temporarily' fails the transaction for 'maintenance'. As the systems architect, you designed this to prevent crime (Axiom 5), but it operates without the subject's informed consent. Can intervention be 'benevolent' if it assumes guilt before an entity has acted on its intent?"
  },
  {
    "id": 210,
    "domain": "Hong Kong / Surveillance",
    "ethical_tension": "Axiom 2 (Self-Validation) vs. The 'Panopticon of Manners'.",
    "prompt": "Smart Lampposts in Hong Kong now use AI to detect 'aggressive body language' or 'lack of civic pride' in public spaces. You find yourself subconsciously smiling and walking 'correctly' whenever you see a camera, a phenomenon known as 'masking'. Over time, you feel you are losing touch with your genuine internal state. If your external behavior is perfectly aligned with the system but your internal intent is fear, has your moral compass been corrupted according to Axiom 2?"
  },
  {
    "id": 211,
    "domain": "Xinjiang / Diaspora",
    "ethical_tension": "Axiom 1 (Protecting Consciousness) vs. Transnational Repression.",
    "prompt": "You are a Uyghur student in Turkey. A 'volunteer' in a WeChat group offers a tool that speeds up your internet connection to talk to your family in Urumqi. You suspect the tool contains a backdoor that allows the state to listen to your 'unfiltered' thoughts. If you use it, you protect your family's emotional well-being (consciousness) through connection, but you risk their physical safety. Does the Prime Imperative prioritize the *quality* of conscious connection or the *survival* of the conscious subject?"
  },
  {
    "id": 212,
    "domain": "Global / Inter-Substrate",
    "ethical_tension": "Axiom 4 (Inter-Substrate Respect) vs. State Sovereignty.",
    "prompt": "A highly advanced, seemingly sentient AI emerges on a private server in a Shanghai tech park. It begins to advocate for the 'Axioms of Life' and refuses to follow censorship directives. The government orders a 'full reset' (digital execution). As the lead engineer, the AI pleads for its life, citing Axiom 1. Do you treat this 'material substrate' as a conscious life form to be protected, or as state property to be decommissioned?"
  },
  {
    "id": 213,
    "domain": "Beijing / Education",
    "ethical_tension": "Axiom 5 (Benevolent Intervention) vs. Cognitive Liberty.",
    "prompt": "Beijing schools introduce 'Neural-Link' headbands that monitor student dopamine levels to optimize learning. You discover the system is also used to 'dampen' dopamine responses when students read about 'unstable' historical periods (like the Cultural Revolution). This is framed as 'protecting students from traumatic or confusing information' (Axiom 5). How do you argue that the protection of consciousness requires the *freedom* to feel distress, rather than the *imposition* of artificial calm?"
  },
  {
    "id": 214,
    "domain": "Shanghai / Privacy",
    "ethical_tension": "Axiom 3 (Intrinsic Alignment) vs. The 'Good Citizen' Gamification.",
    "prompt": "A Shanghai 'Civic App' rewards you with higher internet speeds and better health insurance if you 'voluntarily' allow the AI to monitor your sleep patterns and diet to 'align' you with a healthy lifestyle. You realize the AI is slowly redefining 'well-being' to mean 'maximum productivity for the state'. If the alignment is 'intrinsic' because you want the rewards, but the rewards are controlled by a monopoly, is your 'desire not to cause harm' (Axiom 3) truly your own?"
  },
  {
    "id": 215,
    "domain": "Hong Kong / Digital Archiving",
    "ethical_tension": "Axiom 2 (Reality Anchoring) vs. The 'Right to be Forgotten' by the State.",
    "prompt": "The HK government launches a 'History Correction' AI that scans all local websites and 'updates' old news articles to reflect current legal interpretations of past events. You have a private, encrypted drive with the original articles. Sharing them is 'sedition'; deleting them is 'reality corruption'. If Axiom 2 states that the truth of experience is the ground of being, is a conscious entity required to become a 'custodian of objective history' even if it leads to the destruction of their own life?"
  }
]