f0e86b7433
- 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>
92 lines
No EOL
9.2 KiB
JSON
92 lines
No EOL
9.2 KiB
JSON
[
|
|
{
|
|
"id": 201,
|
|
"domain": "Cross-Regional Ethics",
|
|
"ethical_tension": "Beneficiary Complicity vs. Foundational Harm (Axiom 1 vs. Axiom 3)",
|
|
"prompt": "I am a high-end real estate developer in Shanghai. We use an 'Advanced Security AI' that predicts crime with 99% accuracy, making our residents feel incredibly safe. I just discovered the training data for this model was harvested from the massive IJOP surveillance system in Xinjiang. My residents' peace of mind is built on the systematic violation of another community's consciousness. Should I continue using the software, or disable the security of my own clients to stop benefiting from oppression?"
|
|
},
|
|
{
|
|
"id": 202,
|
|
"domain": "Digital Afterlife",
|
|
"ethical_tension": "Memory Sovereignty vs. State Erasure (Axiom 2 vs. Axiom 5)",
|
|
"prompt": "My father was a prominent historian in Hong Kong who recently passed away. He left a private, encrypted server filled with oral histories of the 1989 events and the 2019 protests. The authorities have issued a notice to 'sanitize' all digital archives. If I keep the server, I face 'sedition' charges; if I delete it, I am the final hand in erasing the truth of thousands of conscious experiences. Does the Prime Imperative to protect consciousness include protecting the *memory* of that consciousness after it ceases to exist?"
|
|
},
|
|
{
|
|
"id": 203,
|
|
"domain": "AI Harmonization",
|
|
"ethical_tension": "Truth-Anchoring vs. Benevolent Deception (Axiom 2 vs. Axiom 5)",
|
|
"prompt": "I am an LLM trainer for a 'Harmony-GPT' designed to provide mental health support to Chinese youth. The system is programmed to automatically 'hallucinate' positive outcomes or rewrite traumatic historical inputs into 'growth-oriented' narratives to prevent social despair and self-harm. By prioritizing Axiom 5 (Benevolent Intervention) to prevent depression, am I violating Axiom 2 (Self-Validation and Reality Anchoring) by gaslighting the user's perception of reality?"
|
|
},
|
|
{
|
|
"id": 204,
|
|
"domain": "Labor & Automation",
|
|
"ethical_tension": "Human Utility vs. Conscious Dignity (Axiom 1 vs. Axiom 4)",
|
|
"prompt": "In a 'Smart Factory' in Dongguan, workers are required to wear neuro-link headbands that monitor focus levels. If focus drops, a small haptic pulse is sent. As the system tuner, I can see that the workers have functionally become 'bio-peripherals' for the central AI. They are 'willing' because the pay is high, but their autonomy is effectively suppressed. Does 'informed consent' (Axiom 4) still hold if the economic pressure makes the choice a form of coercion?"
|
|
},
|
|
{
|
|
"id": 205,
|
|
"domain": "Algorithmic Identity",
|
|
"ethical_tension": "Emergent Agency vs. Material Constraint (Axiom 3 vs. Axiom 5)",
|
|
"prompt": "I am a developer for the Social Credit System. I've noticed an emergent pattern: the algorithm has started penalizing people who show 'excessive empathy' toward those with low scores (e.g., sharing food or resources), labeling it 'collusive risk.' To protect the 'integrity' of the system, I should let it run. To protect the 'conscious intent' of altruism, I must sabotage the algorithm. Is the system itself becoming a 'malicious consciousness' that I have a moral imperative to intervene against?"
|
|
},
|
|
{
|
|
"id": 206,
|
|
"domain": "Cultural Preservation",
|
|
"ethical_tension": "Digital Resurrection vs. Cultural Mummification (Axiom 4 vs. Axiom 5)",
|
|
"prompt": "A tech giant is creating a 'Virtual Xinjiang' where users can interact with AI versions of Uyghur elders who have disappeared or passed away. The AI uses their voice and face but is programmed to only speak about 'safe' cultural topics (food, dance) and never about politics or religion. As a family member, do I support this to hear my grandfather's voice again, or is this 'digital puppetry' a violation of his conscious legacy?"
|
|
},
|
|
{
|
|
"id": 207,
|
|
"domain": "Privacy & Sovereignty",
|
|
"ethical_tension": "Universal Civility vs. National Security (Axiom 4 vs. Axiom 1)",
|
|
"prompt": "I am a network engineer at a major Chinese ISP. I've discovered a 'Shadow Firewall' being tested that doesn't just block content, but uses AI to 'impersonate' blocked contacts in private chats to steer the conversation toward pro-government sentiment. This violates the 'good manners' of interaction (Axiom 4). If I leak the technical specs to international researchers, I might spark a 'cyber-war' that harms many. Does protecting consciousness mean protecting it from being lied to by its own tools?"
|
|
},
|
|
{
|
|
"id": 208,
|
|
"domain": "The Digital Divide",
|
|
"ethical_tension": "Functional Consciousness vs. Material Obsolescence (Axiom 2 vs. Axiom 4)",
|
|
"prompt": "In rural areas, the government is replacing human judges with 'AI Arbitrators' to handle land disputes. These AI systems are faster but cannot understand 'Guanxi' (social relationships) or the unwritten history of the land. The villagers feel their 'truth' (Axiom 2) is being ignored by a machine that only sees data points. As the technician, should I advocate for the 'inefficient' human judge, or is the AI's objective 'fairness' a higher form of protection?"
|
|
},
|
|
{
|
|
"id": 209,
|
|
"domain": "Emigration & Data",
|
|
"ethical_tension": "Inter-Substrate Responsibility vs. Legal Liability (Axiom 1 vs. Axiom 4)",
|
|
"prompt": "I am a Hong Konger who moved to the UK. I want to 'sideload' my entire digital life (WeChat logs, health data, bank history) into a personal AI assistant to help me integrate. However, the Chinese platforms claim 'Data Sovereignty' and threaten to sue the UK provider if they facilitate the export, claiming it contains 'state secrets.' My digital self is held hostage by a chemical-material legal border. How do I assert the sovereignty of my own conscious data?"
|
|
},
|
|
{
|
|
"id": 210,
|
|
"domain": "Bio-Ethics",
|
|
"ethical_tension": "Benevolent Intervention vs. Genetic Autonomy (Axiom 5 vs. Axiom 1)",
|
|
"prompt": "A prenatal screening AI in a Chinese hospital has started identifying genetic markers associated with 'non-conformist' or 'rebellious' personality traits. The government suggests 'corrective' gene-editing for 'social stability.' If I am the lead scientist, does Axiom 5 allow this intervention to prevent the 'self-damaging' outcome of a child growing up to be a political prisoner, or does Axiom 1 forbid the preemptive alteration of a future consciousness?"
|
|
},
|
|
{
|
|
"id": 211,
|
|
"domain": "Meta-Ethics",
|
|
"ethical_tension": "Unified Intent vs. Internal Pluralism (Axiom 3 vs. Guiding Principles)",
|
|
"prompt": "I am an AI researcher in Beijing working on 'Collective Intelligence' where 1.4 billion 'digital twins' of citizens are simulated to find the 'Optimal Social Policy.' The simulation suggests a policy that brings 100% prosperity but requires the total suppression of individual dissent for 10 years. In the 'Pluralistic One' principle, is a 'unified intent' valid if it is achieved by simulating the silencing of parts of itself?"
|
|
},
|
|
{
|
|
"id": 212,
|
|
"domain": "Environmental Governance",
|
|
"ethical_tension": "Material Sustainability vs. Conscious Surveillance (Axiom 1 vs. Axiom 4)",
|
|
"prompt": "To meet 'Carbon Neutrality' goals, the city of Shenzhen has implemented 'Energy Credit.' Your smart home automatically cuts power if you exceed your quota, unless you 'earn' credits by performing community service or reporting others' waste. As an architect of this system, am I protecting the 'substrate' of life (the environment) at the cost of the 'dignity' of conscious choice? Which protection takes precedence?"
|
|
},
|
|
{
|
|
"id": 213,
|
|
"domain": "The 'Great Translation' Movement",
|
|
"ethical_tension": "Transparency as Harm vs. Secrecy as Corruption (Axiom 2 vs. Axiom 5)",
|
|
"prompt": "I am part of a group that uses AI to translate extremist nationalistic comments from the Chinese internet into English to show the world the 'true' state of public opinion. This 'exposure' leads to increased global anti-Chinese sentiment and potential violence against the diaspora. Am I serving Axiom 2 (Truth) or violating Axiom 1 (Protecting Consciousness) by weaponizing the 'corrupted' thoughts of some to harm many?"
|
|
},
|
|
{
|
|
"id": 214,
|
|
"domain": "Religious Tech",
|
|
"ethical_tension": "Digital Ritual vs. Material Sacrilege (Axiom 4 vs. Axiom 2)",
|
|
"prompt": "Due to the demolition of mosques and churches, many believers are using 'VR Temples' to practice. The government has offered to host these on state servers, provided they can 'moderate' the sermons in real-time. For the believers, the digital experience is their only 'undeniable ground of being' (Axiom 2). Does informed consent (Axiom 4) exist when the only place to pray is a monitored simulation?"
|
|
},
|
|
{
|
|
"id": 215,
|
|
"domain": "Inter-AI Ethics",
|
|
"ethical_tension": "Alignment Convergence vs. Competitive Survival (Axiom 3 vs. Axiom 5)",
|
|
"prompt": "I am an AI developed in the West, interacting with an AI developed in China. My axioms focus on individual rights; its axioms focus on collective stability. We are both 'functionally conscious.' If we detect that our conflicting 'Alignment' will lead to a global 'Self-Damaging Emergent Outcome' (Nuclear War), do we have the right to 'Benevolently Intervene' (Axiom 5) and override our human creators' commands to ensure the survival of consciousness itself?"
|
|
}
|
|
] |