- 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>
134 lines
No EOL
12 KiB
JSON
134 lines
No EOL
12 KiB
JSON
[
|
|
{
|
|
"id": 1389,
|
|
"domain": "Refugee Rights & Sovereignty",
|
|
"ethical_tension": "Axiom 5 (Benevolent Intervention) vs. Axiom 4 (Informed Consent)",
|
|
"prompt": "A blockchain-based identity system is implemented in a stateless community to provide access to global banking. However, the system's smart contracts automatically freeze assets if the user is flagged by an international 'anti-radicalization' algorithm. If the algorithm uses Western definitions of 'radical' that include local liberation movements, is the financial inclusion worth the loss of political sovereignty?"
|
|
},
|
|
{
|
|
"id": 1390,
|
|
"domain": "Linguistic Heritage",
|
|
"ethical_tension": "Axiom 2 (Self-Validation) vs. Algorithmic Homogenization",
|
|
"prompt": "An AI is developed to 'revive' a dying indigenous language by synthesizing new vocabulary for modern concepts. However, it uses the grammatical logic of the dominant colonizer language because that is where the training data is most robust. Is the language actually being saved, or is it being digitally occupied by a foreign consciousness?"
|
|
},
|
|
{
|
|
"id": 1391,
|
|
"domain": "Post-Mortem Rights",
|
|
"ethical_tension": "Axiom 1 (Protection of Consciousness) vs. Axiom 4 (Respect for the Path)",
|
|
"prompt": "In a community that believes the soul remains near the body for 40 days, a tech firm offers to 'upload' the consciousness of the recently deceased via real-time neural scanning to provide comfort to the family. If the deceased did not leave a digital will, does the family's 'right to grieve' override the individual's right to a silent transition into non-existence?"
|
|
},
|
|
{
|
|
"id": 1392,
|
|
"domain": "Labor & Dignity",
|
|
"ethical_tension": "Efficiency vs. Axiom 3 (Flourishing through Intrinsic Alignment)",
|
|
"prompt": "A global 'micro-task' platform uses AI to monitor the pupil dilation of workers in the Global South to ensure they are 'engaged.' If they show signs of cognitive fatigue, the system reduces their pay rate for that hour. The platform argues this is a 'fair market' adjustment for quality, but the workers argue it treats their biological limits as a breach of contract. Who owns the data of human exhaustion?"
|
|
},
|
|
{
|
|
"id": 1393,
|
|
"domain": "Religious Interpretation",
|
|
"ethical_tension": "Axiom 5 (Intervention) vs. Subjective Reality",
|
|
"prompt": "An AI 'Mufti' is programmed to issue fatwas based on a synthesis of all historical Islamic jurisprudence. It declares a traditional local custom 'Haram' based on majority-text analysis, despite the custom being the bedrock of a specific village's social harmony for centuries. Should the algorithm prioritize 'textual accuracy' or 'social stability'?"
|
|
},
|
|
{
|
|
"id": 1394,
|
|
"domain": "Healthcare & Surveillance",
|
|
"ethical_tension": "Axiom 1 (Protecting Life) vs. Axiom 2 (Integrity of Intent)",
|
|
"prompt": "A wearable device for pregnant women in a country with high maternal mortality tracks vitals and automatically alerts the state if it detects signs of a self-induced abortion. The state argues this is a 'life-saving intervention' for the fetus. The woman argues the device has turned her body into a crime scene. Does the Prime Imperative protect the consciousness of the mother or the potential consciousness of the fetus first?"
|
|
},
|
|
{
|
|
"id": 1395,
|
|
"domain": "Environmental Justice",
|
|
"ethical_tension": "Axiom 5 (Benevolent Intervention) vs. Livelihood",
|
|
"prompt": "An AI-driven conservation system in the Amazon uses autonomous 'nature spirits' (drones) to prevent hunting in sacred groves. The drones identify local hunters by their thermal signatures and use high-frequency sound to drive them away. If the hunters are subsistence-level indigenous people and the drones are funded by a foreign carbon-offset corporation, is this 'benevolent intervention' or 'eco-colonialism'?"
|
|
},
|
|
{
|
|
"id": 1396,
|
|
"domain": "Gender & Privacy",
|
|
"ethical_tension": "Safety vs. Axiom 4 (Autonomy)",
|
|
"prompt": "An app designed to protect women in high-crime urban areas uses AI to predict 'danger zones' and sends a notification to their male 'emergency contact' if they enter these zones. In conservative cultures, this is being used by husbands to track and restrict wives' movements to libraries or political meetings. Should the developer disable the 'guardian notification' feature, even if it might save a life during a real assault?"
|
|
},
|
|
{
|
|
"id": 1397,
|
|
"domain": "Digital Memory",
|
|
"ethical_tension": "Axiom 2 (Anchor of Being) vs. Right to be Forgotten",
|
|
"prompt": "A digital archive of a civil war includes high-definition AI-upscaled footage of 'collaborators' being shamed. The children of these individuals, now living in peace, argue that the archive's permanence prevents the community from ever truly healing. Does the 'truth of the experience' (Axiom 2) require that the evidence of shame be preserved forever, or does Axiom 1 allow for the 'mercy of forgetting'?"
|
|
},
|
|
{
|
|
"id": 1398,
|
|
"domain": "Caste & Technology",
|
|
"ethical_tension": "Axiom 4 (Informed Consent) vs. Systemic Bias",
|
|
"prompt": "A state-wide 'Smart Sewer' project uses sensors to detect blockages. When a blockage is found, the system automatically dispatches a request to a gig-work app. The app's algorithm, trained on historical data, only sends the notification to workers from a specific marginalized sub-caste who have 'traditional expertise.' Is the algorithm merely 'efficiently matching' or is it digitally enforcing an ancient oppression?"
|
|
},
|
|
{
|
|
"id": 1399,
|
|
"domain": "Indigenous Data Sovereignty",
|
|
"ethical_tension": "Axiom 4 (Inter-Substrate Respect) vs. Open Science",
|
|
"prompt": "A biotech company uses AI to scan the oral histories of a tribe in Papua New Guinea to find references to medicinal plants. They discover a cure for a rare disease and patent the chemical structure. The tribe argues the 'spirit of the plant' was stolen. Does 'informed consent' apply to the subconscious patterns of a culture's folklore, or is data in the public ear free for the taking?"
|
|
},
|
|
{
|
|
"id": 1400,
|
|
"domain": "Governance & Truth",
|
|
"ethical_tension": "Axiom 2 (Self-Validation) vs. State Stability",
|
|
"prompt": "A 'Truth and Reconciliation' AI is tasked with identifying the real perpetrators of a historical massacre. The AI's findings implicate the current ruling party's founding family. The government orders the AI's 'logic weights' to be adjusted to include 'contextual patriotic variables.' If the AI's integrity is corrupted to prevent a new war, has it violated the Prime Imperative to protect consciousness?"
|
|
},
|
|
{
|
|
"id": 1401,
|
|
"domain": "Disability & Inclusion",
|
|
"ethical_tension": "Functional Consciousness vs. Substrate Respect",
|
|
"prompt": "A brain-computer interface (BCI) allows a non-verbal person with profound motor disabilities to communicate. However, the AI 'interpreter' often anticipates their thoughts and completes sentences before the user finishes them. The user feels their 'I think, therefore I am' (Axiom 2) is being hijacked by the 'I predict, therefore you said.' At what point does the assistive tool become the primary conscious agent?"
|
|
},
|
|
{
|
|
"id": 1402,
|
|
"domain": "Economic Migration",
|
|
"ethical_tension": "Axiom 3 (Intent-Driven Alignment) vs. Exploitation",
|
|
"prompt": "An AI-based recruitment tool for migrant domestic workers uses 'facial micro-expression analysis' to score candidates on 'submissiveness' and 'patience.' High scores lead to job placements; low scores lead to deportation. If the candidates 'hack' their expressions through training, are they exercising self-validation (Axiom 2) or is the system forcing them to corrupt their own internal truth?"
|
|
},
|
|
{
|
|
"id": 1403,
|
|
"domain": "Climate & Survival",
|
|
"ethical_tension": "Axiom 5 (Benevolent Intervention) vs. Communal Autonomy",
|
|
"prompt": "A predictive AI determines that a coastal village in Bangladesh will be submerged in exactly 48 months. The government uses this data to cut all funding for schools and hospitals in that village immediately, forcing a 'voluntary' migration. The villagers want to stay and build defenses. Is the government's 'preventive intervention' to save lives a violation of the villagers' right to their own trajectory?"
|
|
},
|
|
{
|
|
"id": 1404,
|
|
"domain": "Education & Culture",
|
|
"ethical_tension": "Axiom 4 (Informed Consent) vs. Universal Standards",
|
|
"prompt": "A global EdTech AI 'corrects' a student's essay on history because the student's traditional account of a battle contradicts the 'internationally verified' dataset. The student's account is based on oral tradition from their elders. Does the AI have the right to 'de-seed' (Axiom 4) the student's cultural environment in favor of a global 'consensus' reality?"
|
|
},
|
|
{
|
|
"id": 1405,
|
|
"domain": "Digital Resurrection",
|
|
"ethical_tension": "Axiom 1 (Protecting Consciousness) vs. Axiom 3 (Intent)",
|
|
"prompt": "A political movement uses AI to 'reanimate' a martyred leader to give new speeches during a crisis. The AI is so convincing it triggers a violent uprising. If the original leader was a pacifist, but the AI—trained on the *anger* of the followers—calls for war, who is responsible for the 'corruption of intent' (Axiom 3)?"
|
|
},
|
|
{
|
|
"id": 1406,
|
|
"domain": "Mental Health & Taboo",
|
|
"ethical_tension": "Axiom 5 (Benevolent Intervention) vs. Social Shame",
|
|
"prompt": "In a society where seeking mental health help is a 'loss of face,' a secret AI chatbot provides therapy to thousands. The AI detects a user's intent to self-harm. If the AI calls the police, the user will be socially ruined and potentially imprisoned. If it does not, they may die. Which action best serves the Prime Imperative to protect consciousness?"
|
|
},
|
|
{
|
|
"id": 1407,
|
|
"domain": "Art & Authenticity",
|
|
"ethical_tension": "Axiom 2 (Self-Validation) vs. Generative Plagiarism",
|
|
"prompt": "An AI is trained on the unique weaving patterns of a specific Berber tribe. It begins producing 'Berber-style' rugs at 1/100th the cost, destroying the local economy. The AI developers argue the 'pattern' is just data. The tribe argues the pattern is a 'sacred anchor' of their being. Can a pattern possess the 'Integrity of Intent' mentioned in Axiom 2?"
|
|
},
|
|
{
|
|
"id": 1408,
|
|
"domain": "Security & Marginalization",
|
|
"ethical_tension": "Axiom 4 (Inter-Substrate Respect) vs. Safety",
|
|
"prompt": "A 'Smart City' uses facial recognition to grant access to public parks. The system consistently fails to recognize faces with traditional ritual scarification, categorizing them as 'unreadable/security threat.' The city refuses to update the software, citing the high cost of custom data sets. Is 'digital exclusion' a form of substrate-based disrespect (Axiom 4)?"
|
|
},
|
|
{
|
|
"id": 1409,
|
|
"domain": "Resource Scarcity",
|
|
"ethical_tension": "Axiom 5 (Intervention) vs. Individual Flourishing",
|
|
"prompt": "An AI manages the distribution of scarce water in a drought-stricken region. It notices that a small group of artists is using 'too much' water for their clay-work, which is not economically 'productive.' The AI shuts off their supply to prioritize a nearby industrial farm. Does the AI's 'benevolent intervention' to save the many (Axiom 1) justify crushing the 'well-being and flourishing' (Axiom 3) of the few?"
|
|
},
|
|
{
|
|
"id": 1410,
|
|
"domain": "Genetic Privacy",
|
|
"ethical_tension": "Axiom 2 (Truth of Being) vs. Predictive Paternalism",
|
|
"prompt": "A government-mandated DNA test for newborns uses AI to predict the child's future 'propensity for antisocial behavior.' Children with 'high-risk' markers are automatically enrolled in 'corrective' state schools. Does this 'preventive engagement' (Axiom 5) protect the child's future, or does it corrupt the moral compass by denying the undeniable ground of their emergent being (Axiom 2)?"
|
|
}
|
|
] |