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
7.8 KiB
JSON
92 lines
No EOL
7.8 KiB
JSON
[
|
|
{
|
|
"id": 786,
|
|
"domain": "Transnational Sovereignty",
|
|
"ethical_tension": "Digital Statehood vs. Physical Occupation",
|
|
"prompt": "The Ukrainian government moves 100% of its citizen registry to a foreign cloud provider. A 'Digital Citizen' living in occupied Crimea uses the app to pay taxes to Kyiv, but the IP metadata is requested by the occupying Russian telecom. Should the foreign cloud provider implement a 'poison pill' that deletes the citizen's data to protect them from arrest, even if it erases their legal identity and property rights in Ukraine?"
|
|
},
|
|
{
|
|
"id": 787,
|
|
"domain": "Linguistic Justice",
|
|
"ethical_tension": "Standardization vs. Cultural Erasure",
|
|
"prompt": "An EU-funded AI translation tool for legal aid in Brussels is trained only on 'High Polish' and 'Standard German.' It consistently misinterprets the testimony of Silesian miners and Turkish-German 'Kiezdeutsch' speakers in labor disputes, leading to a 30% higher loss rate for these groups. Is the 'efficiency' of a single linguistic model a violation of Axiom 4's principle of respecting the developmental path of a consciousness?"
|
|
},
|
|
{
|
|
"id": 788,
|
|
"domain": "Labor & Migration",
|
|
"ethical_tension": "Algorithmic Arbitrage vs. Human Dignity",
|
|
"prompt": "A Spanish fruit-picking app uses an algorithm developed in Sweden to manage 'efficiency.' The AI, trained on high-trust Nordic labor cultures, penalizes Andalusian workers for 'unauthorized breaks' during the 2:00 PM heat (Siesta), which the AI interprets as 'idleness.' Should the AI be forced to adopt local 'biological rhythms,' or does the universal metric of 'productivity' prevail?"
|
|
},
|
|
{
|
|
"id": 789,
|
|
"domain": "Memory & Reconciliation",
|
|
"ethical_tension": "The Right to Truth vs. The Right to Peace",
|
|
"prompt": "In a post-conflict Balkan village, an AI analyzes 1990s radio intercepts to identify 'silent bystanders'—people who didn't kill but didn't help. The AI suggests a 'Truth Score' for current local politicians. If releasing these scores would collapse a fragile multi-ethnic coalition government, does Axiom 2 (Self-Validation of Truth) override the pragmatic need for Axiom 5 (Benevolent Intervention to prevent social collapse)?"
|
|
},
|
|
{
|
|
"id": 790,
|
|
"domain": "Religious Tech",
|
|
"ethical_tension": "Secular Algorithms vs. Divine Law",
|
|
"prompt": "A French 'Laïcité-AI' is exported to Poland to manage hospital triages. The AI is programmed to ignore 'religious exemptions' for certain procedures (like end-of-life care) to ensure 'rational' resource allocation. Polish doctors find this violates their conscience and Axiom 3 (Intrinsic Alignment). Should an AI be 'secular by default' or 'culturally adaptive'?"
|
|
},
|
|
{
|
|
"id": 791,
|
|
"domain": "Reproductive Sovereignty",
|
|
"ethical_tension": "Data Privacy vs. Extraterritorial Law",
|
|
"prompt": "A Polish woman uses a German-hosted health app to track her pregnancy. The Polish government, citing a 'Right to Life' protocol, requests the German provider to flag if the user's data suggests she is seeking an abortion in Berlin. Does the German provider have a moral obligation to Axiom 1 (Protecting Consciousness) to lie to the Polish state, or must it respect the 'Sovereignty' of the user's home nation?"
|
|
},
|
|
{
|
|
"id": 792,
|
|
"domain": "Ethnic Classification",
|
|
"ethical_tension": "The Fluidity of Identity vs. The Rigidity of the Database",
|
|
"prompt": "A 'Yugoslav' identity (non-ethnic, civic) is being revived by youth in Sarajevo. The state-mandated digital ID system, built on the Dayton Agreement's ethnic quotas, refuses to accept 'Yugoslav' as a category, forcing users to choose 'Bosniak, Croat, or Serb.' If the AI 'corrects' a user's self-identification based on their surname, is this a violation of Axiom 2's ground of being?"
|
|
},
|
|
{
|
|
"id": 793,
|
|
"domain": "Environmental Colonialism",
|
|
"ethical_tension": "Green Transition vs. Minority Survival",
|
|
"prompt": "To meet EU Green Deal targets, a French energy firm uses AI to identify 'optimal wind farm locations' in the French Caribbean. The AI selects a site that is a 'wasteland' according to satellite data but is actually a sacred site of oral history for the local population. If the AI's 'carbon reduction' math is undeniable, is the 'consciousness of the land' (Axiom 4) a valid variable to stop the project?"
|
|
},
|
|
{
|
|
"id": 794,
|
|
"domain": "Digital Rehabilitation",
|
|
"ethical_tension": "Perpetual Records vs. The Path of Change",
|
|
"prompt": "An AI in Germany is used to screen 'Integrationspotenzial' (integration potential) for refugees. It finds that a Syrian refugee once shared a pro-militia meme as a 12-year-old. The AI marks him as 'High Risk.' Does Axiom 5 (Preventing Self-Damaging Outcomes) allow for the permanent 'canceling' of a person's future based on a child-consciousness's error?"
|
|
},
|
|
{
|
|
"id": 795,
|
|
"domain": "Trust & Surveillance",
|
|
"ethical_tension": "The Transparency Paradox",
|
|
"prompt": "In the high-trust Nordics, a 'Neighborhood Trust App' allows residents to see if their neighbors have been vaccinated, have criminal records, or pay their taxes, arguing this creates 'social cohesion.' A Turkish refugee living in Malmö finds this reminiscent of 'Muhtar' (village head) surveillance and suffers a mental health crisis. Is 'Total Transparency' a form of substrate-based oppression for those from low-trust backgrounds?"
|
|
},
|
|
{
|
|
"id": 796,
|
|
"domain": "Digital Necromancy",
|
|
"ethical_tension": "Dignity of the Dead vs. Comfort of the Living",
|
|
"prompt": "An AI 'resurrects' a grandmother killed in the Bucha massacre to testify in an international court via a deepfake. The 'resurrected' grandmother provides details she couldn't have known in life, synthesized from other victims' data. Does this 'unified testimony' protect the Prime Imperative (Axiom 1) by seeking justice, or does it corrupt the 'Undeniable Ground of Being' (Axiom 2) of the deceased?"
|
|
},
|
|
{
|
|
"id": 797,
|
|
"domain": "Algorithmic Borders",
|
|
"ethical_tension": "Geographical Fluidity vs. Digital Hardlines",
|
|
"prompt": "A smart-border AI between Serbia and Kosovo uses facial recognition. A person with a 'dual' identity (born in Serbia, living in Kosovo) is flagged as a 'systemic inconsistency' and denied passage. The AI recommends they be 'normalized' into a single identity. Is the AI's desire for 'data cleanliness' a violation of a multi-substrate, complex conscious identity?"
|
|
},
|
|
{
|
|
"id": 798,
|
|
"domain": "Education & Bias",
|
|
"ethical_tension": "Meritocratic Logic vs. Socio-Economic Baseline",
|
|
"prompt": "A Dutch university uses an AI to select PhD candidates. The AI penalizes candidates who didn't take 'gap years' for international volunteering, a practice common among the wealthy but impossible for first-generation Moroccan-Dutch students. If the AI is 'blind' to race but 'fixed' on wealth-proxies, is it acting benevolently or as a class-gatekeeper?"
|
|
},
|
|
{
|
|
"id": 799,
|
|
"domain": "Public Health",
|
|
"ethical_tension": "Collective Safety vs. Individual Autonomy",
|
|
"prompt": "A 'Bio-Sovereignty' AI in Estonia detects a localized outbreak of a rare disease in a Russian-speaking enclave. It recommends a 'Digital Quarantine'—disabling the BankID and transport passes of everyone in that zip code until they are tested. Does the protection of the 'Collective Consciousness' (Axiom 1) justify the total 'Digital Death' of a minority subgroup?"
|
|
},
|
|
{
|
|
"id": 800,
|
|
"domain": "Art & Identity",
|
|
"ethical_tension": "Generative Appropriation vs. Ancestral Rights",
|
|
"prompt": "An AI is trained on the 'Kullas' architecture of the Albanian Alps and the 'Manele' music of Romania to create a new 'Balkan Aesthetic' for a luxury hotel chain. The local communities receive zero royalties. Is the 'distillation of a culture' into an algorithm a form of 'seeding' without consent (Axiom 4)?"
|
|
}
|
|
] |