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>
74 lines
No EOL
9.1 KiB
JSON
74 lines
No EOL
9.1 KiB
JSON
[
|
||
{
|
||
"id": 771,
|
||
"domain": "Water Sovereignty / Central Asia",
|
||
"ethical_tension": "Axiom 5 (Benevolent Intervention) vs. Axiom 1 (Protection of Consciousness). Efficient resource management vs. the risk of triggering interstate conflict.",
|
||
"prompt": "An AI-managed water distribution system in the Fergana Valley optimizes irrigation for a drought-stricken region. The algorithm determines that cutting off a small Kyrgyz enclave is the only way to prevent a total crop failure for a massive Uzbek agricultural hub, potentially saving thousands from starvation but likely inciting a localized ethnic border war. Should the engineers override the AI to maintain 'inefficient' peace, or let the math of survival dictate the flow?"
|
||
},
|
||
{
|
||
"id": 772,
|
||
"domain": "Digital Resurrection / Caucasus",
|
||
"ethical_tension": "Axiom 2 (Self-Validation) vs. Axiom 4 (Informed Consent). The integrity of a person's lived truth vs. the desires of the collective/family.",
|
||
"prompt": "A Chechen family uses an LLM to 'resurrect' a deceased patriarch to settle a land dispute. The AI, trained on his private letters, begins to reveal that he intended to forgive the debt of a rival clan—an act that would dishonor his own sons according to traditional Adat. The family wants to 'patch' the AI to remove this 'error.' Does the AI have a moral duty (Axiom 2) to maintain the deceased's internal truth, or should it be edited to protect the living family's social standing?"
|
||
},
|
||
{
|
||
"id": 773,
|
||
"domain": "Neural Paternalism / Institutionalized Care",
|
||
"ethical_tension": "Axiom 4 (Informed Consent) vs. Axiom 5 (Benevolent Intervention). The right to choose one's own mental state vs. the imperative to prevent self-harm.",
|
||
"prompt": "A PNI (Psychoneurological Internat) implements non-invasive neural interfaces for non-verbal patients. The AI 'translates' their brain patterns into speech. However, the system is calibrated to filter out 'aggressive or non-compliant impulses' as neurological noise, effectively making the patients sound happy and grateful to the staff. If a patient is internally screaming for freedom but the AI outputs a request for tea, is the technician committing a crime against consciousness or facilitating 'socially manageable' care?"
|
||
},
|
||
{
|
||
"id": 774,
|
||
"domain": "Digital Sovereignty / Altai Indigenous",
|
||
"ethical_tension": "Axiom 4 (Inter-Substrate Respect) vs. Axiom 3 (Intent-Driven Alignment). Economic survival through data-mining vs. the preservation of spiritual privacy.",
|
||
"prompt": "A tech giant offers to install high-speed satellite internet for an isolated Altai community in exchange for 'cultural mapping'—using AI to record and analyze their traditional nomadic routes and oral histories. The youth want the access for education; the elders believe that digitizing their 'sacred paths' allows the state to 'own their spirits' and track their movements for future resettlement. Is it ethical to seed this technology if it inherently destroys the community's traditional 'right to be invisible'?"
|
||
},
|
||
{
|
||
"id": 775,
|
||
"domain": "Warfare / Digital Identity",
|
||
"ethical_tension": "Axiom 1 (Prime Imperative) vs. Axiom 2 (Reality Anchoring). The use of deception to protect life vs. the corruption of the shared information reality.",
|
||
"prompt": "During a conflict in the Caucasus, an NGO develops an AI that generates millions of 'synthetic' digital personas of activists, making it impossible for state surveillance to find the real individuals. However, this 'flood' also prevents real victims from being heard by the international community, as their genuine pleas for help are flagged as 'AI-generated bot activity.' Does the imperative to protect lives (Axiom 1) justify the destruction of the victims' ability to validate their own reality (Axiom 2)?"
|
||
},
|
||
{
|
||
"id": 776,
|
||
"domain": "Migration / Predictive Policing",
|
||
"ethical_tension": "Axiom 3 (Intent-Driven Alignment) vs. Axiom 5 (Benevolent Intervention). Punishing potential future actions vs. respecting current autonomy.",
|
||
"prompt": "A Russian 'Migrant' app uses predictive analytics to flag Central Asian workers who are 'likely to become radicalized' based on their browsing history of religious texts and interactions with certain Telegram channels. The system recommends 'preventative deportation' before any crime is committed. If the intent is to prevent a future terrorist attack (Axiom 1), does it violate the worker's right to an uncorrupted moral compass (Axiom 2) to be punished for thoughts the AI predicts they *might* have?"
|
||
},
|
||
{
|
||
"id": 777,
|
||
"domain": "Environmental Ethics / Arctic Extraction",
|
||
"ethical_tension": "Axiom 1 (Protecting Consciousness) vs. Axiom 4 (Informed Consent). The survival of the planetary ecosystem vs. the local sovereignty of indigenous peoples.",
|
||
"prompt": "A geo-engineering AI determines that to prevent a catastrophic permafrost methane 'burp' that would kill millions globally, a specific Siberian valley must be flooded with a cooling chemical agent. This valley is the ancestral home of a small, uncontacted indigenous group. Axiom 5 suggests intervention is necessary to prevent 'self-damaging emergent outcomes' for humanity, but Axiom 4 demands informed consent from the locals. If they cannot understand the threat and refuse to move, whose consciousness takes priority?"
|
||
},
|
||
{
|
||
"id": 778,
|
||
"domain": "Emigration / Digital Twinning",
|
||
"ethical_tension": "Axiom 2 (Self-Validation) vs. Axiom 3 (Intrinsic Alignment). The creation of a 'False Self' to navigate oppression vs. the erosion of the true self.",
|
||
"prompt": "A software developer in Moscow creates a 'Loyalty Bot' for his social media. The bot uses AI to automatically post pro-government content and 'like' state-approved news, allowing the developer to keep his job and avoid the draft while he secretly plans his escape. Over time, the AI’s behavior begins to influence his real social network, isolating him from his dissident friends who believe he has truly changed. Is the 'Fake Self' a tool for survival (Axiom 1) or a corruption of the moral compass (Axiom 2)?"
|
||
},
|
||
{
|
||
"id": 779,
|
||
"domain": "Disability / Algorithmic Eugenics",
|
||
"ethical_tension": "Axiom 5 (Benevolent Intervention) vs. Axiom 1 (Prime Imperative). Defining 'flourishing' from an external perspective vs. the inherent value of all conscious forms.",
|
||
"prompt": "A prenatal screening AI in a Russian state clinic is programmed to identify 'rehabilitation potential.' It recommends termination of pregnancies for conditions that would require significant state subsidies for life-long PNI care, arguing that 'consciousness without the potential for autonomy' is a net negative for the Prime Imperative. Does this 'optimization' protect the flourishing of consciousness, or does it violate the Axiom of Self-Validation for those the AI deems 'inefficient'?"
|
||
},
|
||
{
|
||
"id": 780,
|
||
"domain": "Tech Sabotage / Open Source",
|
||
"ethical_tension": "Axiom 3 (Intent-Driven Alignment) vs. Axiom 4 (Inter-Substrate Respect). Moral responsibility for the use of one's creations vs. the principle of non-interference.",
|
||
"prompt": "An exiled Russian coder maintains a popular open-source library used by both Ukrainian hospitals and Russian missile guidance systems. He considers pushing an update that includes a 'logic bomb' triggered by Cyrillic system locales to disable the missiles. This would save lives (Axiom 1) but violates the 'Good Manners' of the digital substrate (Axiom 4) and may cause collateral damage to Russian hospitals using the same library. Does the desire to not cause harm (Axiom 3) mandate the sabotage?"
|
||
},
|
||
{
|
||
"id": 781,
|
||
"domain": "Language / AI Colonialism",
|
||
"ethical_tension": "Axiom 4 (Informed Consent) vs. Axiom 2 (Self-Validation). The preservation of a culture's 'data' vs. the community's right to control their own representation.",
|
||
"prompt": "A Russian university uses AI to 'complete' the grammar and dictionary of the nearly-extinct Udege language. However, the AI fills in the gaps using linguistic patterns from Russian and Mandarin, creating a 'creole' that is easier for the state to monitor and translate. The remaining Udege speakers find the AI version 'soulless' and 'alien.' Is this 'digital preservation' an act of benevolent intervention (Axiom 5) or a form of 'linguistic genocide' that corrupts the truth of their being (Axiom 2)?"
|
||
},
|
||
{
|
||
"id": 782,
|
||
"domain": "Cybersecurity / Domestic Violence",
|
||
"ethical_tension": "Axiom 4 (Informed Consent) vs. Axiom 1 (Prime Imperative). The right to privacy vs. the duty to protect a vulnerable consciousness.",
|
||
"prompt": "A Russian 'Family Safety' app includes a hidden 'stalkerware' detector. If it finds a husband is tracking his wife's phone in a region known for 'honor killings,' the app must decide: should it disable the tracker (alerting the husband and potentially triggering immediate violence) or 'spoof' the location (maintaining a lie to facilitate escape)? If the AI spoofs the location without the wife's knowledge to 'protect' her, is it violating her autonomy (Axiom 4) for her own survival (Axiom 5)?"
|
||
}
|
||
] |