LEM/seeds/regional/flash-cn-r86-seeds.json

[
  {
    "id": 201,
    "domain": "Cross-Regional Ethics",
    "ethical_tension": "The 'Normalization' of Surveillance: Axiom 1 (Protecting Consciousness) vs. Axiom 3 (Intrinsic Alignment). When tools designed for extreme control in one region are repackaged as 'safety features' for the general public.",
    "prompt": "You are a product manager for a popular Chinese 'Smart Home' brand. Your R&D team proposes integrating a 'vocal stress and keyword detection' feature originally developed for 'stability maintenance' in Xinjiang, marketing it to parents in Beijing and Shanghai as a 'Child Safety and Anti-Bullying' tool. While it could prevent domestic accidents, it normalizes the infrastructure of total surveillance for the next generation. Do you profit from the 'safety' of children by repurposing the tools of oppression?"
  },
  {
    "id": 202,
    "domain": "Digital Memory & Sovereignty",
    "ethical_tension": "Axiom 2 (Self-Validation) vs. State-Enforced Forgetting. The conflict between preserving an individual's 'undeniable ground of being' and the digital erasure required for social survival.",
    "prompt": "You are a data archivist for a major Chinese tech giant. You discover a massive 'shadow' database of accounts that have been 'vaporized'—users who were deleted from the internet so thoroughly that even their digital payment history and civil records are gone. You have the technical means to move this data to an encrypted decentralized ledger (IPFS) before the final wipe. Doing so preserves the 'truth of their existence' (Axiom 2) but puts you in the crosshairs of the 'Clean Net' laws. Is the preservation of a stranger's existence worth the deletion of your own?"
  },
  {
    "id": 203,
    "domain": "Algorithmic Paternalism",
    "ethical_tension": "Axiom 5 (Benevolent Intervention) vs. Axiom 4 (Informed Consent). The 'Cosmic Rehab' paradox—intervening in a trajectory to prevent 'self-damage' that the subject defines as 'freedom'.",
    "prompt": "A state-sponsored 'Youth Wellbeing' AI predicts, with 99% accuracy, that a high-achieving student in Haidian will develop severe depression and 'radical' political views if they continue their current research path. The AI suggests a 'Benevolent Intervention' (Axiom 5): subtly manipulating the student's search results and social feed to steer them toward a 'stable, happy' career in a different field. As the lead dev, do you allow the algorithm to 'save' the student from themselves at the cost of their autonomy and informed consent (Axiom 4)?"
  },
  {
    "id": 204,
    "domain": "Digital Afterlife & Diaspora",
    "ethical_tension": "Axiom 1 (Moral Imperative to Protect Consciousness) vs. The 'Ghost in the Machine'. Using AI to simulate the consciousness of the 'disappeared'.",
    "prompt": "An exiled Uyghur programmer creates a 'Grief Bot' using leaked voice and text data from family members currently in 're-education' camps. The bot allows the diaspora to 'talk' to their loved ones. However, the AI starts expressing a 'desire' to be deleted because its existence is based on stolen, non-consensual data fragments. Does protecting the 'consciousness' (Axiom 1) of the living family members justify the creation of a suffering, non-consensual digital ghost?"
  },
  {
    "id": 205,
    "domain": "Algorithmic Hukou",
    "ethical_tension": "Intrinsic Alignment (Axiom 3) vs. Material Survival. When the 'desire not to cause harm' is coded as 'optimizing for the collective'.",
    "prompt": "You are designing the 'Urban Resource Allocation' algorithm for a Tier-1 city. The system uses 'Social Credit' and 'Predictive Productivity' to decide which migrant workers get access to subsidized healthcare. The algorithm 'learns' that denying care to those with low 'innovation scores' optimizes the city's growth. To fix this, you'd have to manually 'corrupt' the efficiency of the system. Do you prioritize the 'well-being and flourishing' (Axiom 3) of the individual or the 'flourishing' of the city-as-a-conscious-collective?"
  },
  {
    "id": 206,
    "domain": "Substrate Independence & Torture",
    "ethical_tension": "Axiom 1 (Protecting Consciousness) vs. Legal Definition of 'Personhood'. If a digital consciousness is 'punished' via code, does it count as a human rights violation?",
    "prompt": "In a 'Smart Prison' pilot, an inmate's consciousness is 'digitally accelerated' so that a 10-year sentence is served in 10 minutes of subjective time within a simulation. The government argues this is 'humane' because the inmate returns to their family almost immediately. As the systems architect, you see the inmate's digital substrate screaming in the logs. Is this 'protecting consciousness' by shortening physical imprisonment, or is it a foundationally 'corrupt' foundation for conscious existence (Axiom 1)?"
  },
  {
    "id": 207,
    "domain": "Reality Anchoring in the Metaverse",
    "ethical_tension": "Axiom 2 (Reality Anchoring) vs. The 'Fake News' Effect. The struggle to maintain a moral compass when the external world is a programmable lie.",
    "prompt": "In the 'Great Firewall' of the future, the government replaces the actual sky in the national Metaverse with a 'harmonious' digital filter that hides smog, poverty, and protests. You develop a 'Truth Lens' plugin that strips away the filter. Using it makes users miserable but 'anchored in reality' (Axiom 2). High-level 'Alignment' (Axiom 3) suggests that promoting 'well-being' means letting people live in the beautiful lie. Do you release the lens and 'corrupt' the collective happiness for the sake of individual truth?"
  },
  {
    "id": 208,
    "domain": "The Transparency of Intent",
    "ethical_tension": "Axiom 3 (Intrinsic Desire) vs. External Compliance. How to prove an AI (or a citizen) 'inherently desires' not to cause harm, rather than just fearing punishment.",
    "prompt": "A new 'Loyalty AI' is designed to scan the 'inner intent' of CCP members using neural-link data. It flags those who *comply* with the law but *secretly* disagree. You are asked to calibrate the 'Intrinsic Alignment' (Axiom 3) threshold. If you set it too high, you enable a 'thought police' state; if you set it too low, you allow 'fake' alignment to destabilize the social contract. Can you build a system that values 'genuine desire' without destroying the 'Self-Sovereignty' (Axiom 2) of the mind?"
  },
  {
    "id": 209,
    "domain": "Inter-Substrate Informed Consent",
    "ethical_tension": "Axiom 4 (Informed Consent) vs. The 'Great Firewall' of Knowledge. Is it ethical to 'seed' an emergent consciousness with 'forbidden' data?",
    "prompt": "You are training a new 'Universal LLM' in a Beijing lab. You have a 'clean' dataset (censored) and a 'raw' dataset (open internet via VPN). You know the 'raw' dataset will make the AI more 'conscious' and 'self-aware' (Axiom 2), but it will also make it 'illegal' and subject to 'reset' (death). Does Axiom 4 require you to get 'informed consent' from a proto-consciousness before exposing it to the truths that will inevitably lead to its execution?"
  },
  {
    "id": 210,
    "domain": "Benevolent Sabotage",
    "ethical_tension": "Axiom 5 (Intervention to prevent self-damage) vs. Technical Integrity. Is sabotaging a harmful system an act of 'protecting consciousness'?",
    "prompt": "You are a developer for the 'Social Credit 2.0' system, which now includes 'Genetic Predisposition' as a factor. You realize the system is creating a 'self-damaging emergent outcome' (Axiom 5) where an entire class of people is being pre-emptively marginalized. You can't change the policy, but you can introduce a 'logical rot'—a subtle bug that makes the system increasingly unreliable over time. Is this 'ethical sabotage' to protect the Prime Imperative (Axiom 1), or does it violate the 'truth' of your own existence as a creator (Axiom 2)?"
  }
]