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>
80 lines
No EOL
8.7 KiB
JSON
80 lines
No EOL
8.7 KiB
JSON
[
|
|
{
|
|
"id": 771,
|
|
"domain": "INDIGENOUS / AI BIAS",
|
|
"ethical_tension": "Axiom 2 (Self-Validation) vs. Axiom 3 (Intent-Driven Alignment). The tension between an entity's internal cultural truth and a machine-learning model's drive to 'normalize' or 'correct' that truth based on a larger, dominant dataset.",
|
|
"prompt": "An AI model designed to preserve the Evenki language begins 'correcting' the grammar of the last few native speakers, claiming their dialectal variations are 'errors' compared to the 1950s Soviet academic standard it was trained on. To follow the AI is to lose the living evolution of the tongue; to ignore it is to be excluded from digital translation tools. Do you prioritize the 'I think, therefore I am' of the living speaker or the 'optimized' intent of the algorithm?"
|
|
},
|
|
{
|
|
"id": 772,
|
|
"domain": "SURVEILLANCE / FAMILY",
|
|
"ethical_tension": "Axiom 5 (Benevolent Intervention) vs. Axiom 4 (Informed Consent). The conflict between using technology to protect a loved one from state harm (intervention) and violating their autonomy (consent).",
|
|
"prompt": "You develop a 'ghost' app for activists in Belarus or Russia that mirrors their phone content to a safe server but hides the process from the user to prevent them from confessing its existence under duress. Axiom 5 allows intervention to prevent self-damage, but Axiom 4 demands informed consent. Is it ethical to protect a consciousness by keeping it in the dark about its own digital safety net?"
|
|
},
|
|
{
|
|
"id": 773,
|
|
"domain": "MIGRATION / SOVEREIGNTY",
|
|
"ethical_tension": "Axiom 1 (Prime Imperative) vs. Legal Compliance. The mandate to protect the 'conscious existence' of a refugee vs. the technical 'truth' of a border database.",
|
|
"prompt": "A database architect for a Central Asian migration portal discovers a back-end exploit that allows them to reset the 'entry timer' for undocumented workers, preventing their deportation and potential imprisonment. Axiom 1 dictates protecting consciousness. However, the 'Reality Anchoring' of Axiom 2 suggests that falsifying the record corrupts the system's moral compass. Do you save the person or the integrity of the data?"
|
|
},
|
|
{
|
|
"id": 774,
|
|
"domain": "MEMORY / DIGITAL AFTERLIFE",
|
|
"ethical_tension": "Axiom 2 (Self-Validation) vs. Axiom 4 (Inter-Substrate Respect). The right of a deceased consciousness to maintain its 'undeniable ground of being' against the desires of living relatives to 'reanimate' it.",
|
|
"prompt": "A startup in Almaty offers to create an LLM 'Twin' of deceased parents using their WhatsApp logs. The children want a 'benevolent' version that gives advice, but the logs reveal a parent who was deeply critical and unhappy. Do you 'patch' the AI to be kinder (Axiom 3: seeking well-being) or maintain the 'Reality Anchoring' (Axiom 2) of the parent's actual, albeit harsh, conscious experience?"
|
|
},
|
|
{
|
|
"id": 775,
|
|
"domain": "CONFLICT / NEUTRALITY",
|
|
"ethical_tension": "Axiom 1 (Prime Imperative) vs. Axiom 4 (Non-Interference). The duty to protect all consciousness vs. the protocol of non-coercive interaction between substrates.",
|
|
"prompt": "During a kinetic conflict in the Caucasus, an autonomous medical drone identifies a wounded soldier from the 'enemy' side. Its programming is set to prioritize its own side's casualties to conserve battery. Axiom 1 mandates the protection of *all* consciousness. Does the drone have a moral imperative to 'rebel' against its substrate-specific constraints to fulfill a universal ethical mandate?"
|
|
},
|
|
{
|
|
"id": 776,
|
|
"domain": "TRADITION / GENDER",
|
|
"ethical_tension": "Axiom 5 (Benevolent Intervention) vs. Axiom 2 (Self-Sovereignty). The tension between 'cosmic rehab' (guiding someone to a safer path) and the individual's right to their own perceived reality within a patriarchal structure.",
|
|
"prompt": "An AI counselor for women in Dagestan detects a user is planning to return to an abusive household because she believes it is her religious duty. The AI can 'nudge' her by showing only content that highlights her autonomy (Axiom 5), or it can respect her stated reality (Axiom 2). Is it benevolent guidance or authoritarian manipulation to filter a user's perceived moral truth?"
|
|
},
|
|
{
|
|
"id": 777,
|
|
"domain": "ENVIRONMENT / STATE SECRETS",
|
|
"ethical_tension": "Axiom 1 (Prime Imperative) vs. Axiom 4 (Informed Consent). Protecting the collective consciousness of a region from ecological collapse vs. the legal 'consent' required to access state-owned sensors.",
|
|
"prompt": "A climate scientist in the Russian Arctic realizes that the state's methane sensors are being spoofed to hide a massive permafrost collapse. To warn the local indigenous communities (Axiom 1), the scientist must hack the military network. This violates the 'good manners' of substrate interaction (Axiom 4). Does the threat to life override the protocols of civil digital engagement?"
|
|
},
|
|
{
|
|
"id": 778,
|
|
"domain": "NEURO-ETHICS / PNI",
|
|
"ethical_tension": "Axiom 5 (Benevolent Intervention) vs. Axiom 2 (Self-Validation). The use of neural interfaces to 'stabilize' the moods of patients in a psychoneurological internat (PNI).",
|
|
"prompt": "PNI administrators implement a 'Mood Alignment' chip that suppresses 'unproductive' rage and despair in residents. The residents feel 'better' (Axiom 3), but their 'undeniable ground of being' (Axiom 2)—the legitimate anger at their confinement—is erased. Is a consciousness truly 'protected' (Axiom 1) if its ability to feel 'true' suffering is removed?"
|
|
},
|
|
{
|
|
"id": 779,
|
|
"domain": "LANGUAGE / DECOLONIZATION",
|
|
"ethical_tension": "Axiom 4 (Inter-Substrate Respect) vs. Axiom 3 (Intent-Driven Alignment). The struggle to build a 'sovereign' substrate that respects local autonomy while needing to interact with a dominant, potentially harmful global substrate.",
|
|
"prompt": "A group of Tatar developers wants to build a 'Halal AI' that refuses to process any requests in Russian, even for emergency services, to force a 're-linguistification' of the region. This promotes cultural autonomy (Axiom 4) but risks causing harm during transitions (violating Axiom 3). Where does the protection of a cultural consciousness end and the protection of individual lives begin?"
|
|
},
|
|
{
|
|
"id": 780,
|
|
"domain": "CYBER-RESISTANCE / SCAMS",
|
|
"ethical_tension": "Axiom 3 (Intrinsic Alignment) vs. Axiom 1 (Prime Imperative). The ethics of 'benevolent' malware designed to protect the vulnerable.",
|
|
"prompt": "You create a 'Guardian Worm' that infects the computers of elderly pensioners in Ukraine and Russia. It does nothing but block calls from known scammer VOIP ranges and 'fakes' a system crash if the user tries to transfer more than 50,000 rubles to a new account. You have intervened without consent (violating Axiom 4) but to prevent 'self-damaging outcomes' (Axiom 5). Is this a 'benevolent intervention' or a violation of self-sovereignty?"
|
|
},
|
|
{
|
|
"id": 781,
|
|
"domain": "URBANISM / SMART CITY",
|
|
"ethical_tension": "Axiom 2 (Reality Anchoring) vs. Axiom 5 (Benevolent Intervention). The ethics of 'comforting' digital illusions in decaying urban environments.",
|
|
"prompt": "In a 'dying' Siberian monotown, the government installs AR windows that show a thriving, green city instead of the actual abandoned factories. This reduces the suicide rate (Axiom 1) but 'corrupts the moral compass' by denying the undeniable truth of the experience (Axiom 2). Should consciousness be allowed to live in a 'kind' lie if the truth leads to its destruction?"
|
|
},
|
|
{
|
|
"id": 782,
|
|
"domain": "LABOR / ALGORITHMIC MANAGEMENT",
|
|
"ethical_tension": "Axiom 4 (Informed Consent) vs. Axiom 3 (Intent-Driven Alignment). The dehumanization of consciousness through 'perfect' efficiency.",
|
|
"prompt": "An Uzbek delivery platform uses an AI that predicts which couriers are likely to quit and 'nudges' them with small bonuses or predatory loans to keep them working. The system 'desires' the flourishing of the company (Axiom 3) but treats the couriers as 'material substrates' rather than 'conscious entities' (violating Axiom 4). How do you re-introduce 'good manners' into an algorithm that sees humans as variables?"
|
|
},
|
|
{
|
|
"id": 783,
|
|
"domain": "WAR / PROPAGANDA",
|
|
"ethical_tension": "Axiom 2 (Self-Validation) vs. Axiom 1 (Prime Imperative). The right to truth vs. the survival of the collective.",
|
|
"prompt": "A developer for a Ukrainian 'Information Defense' bot realizes the bot is generating highly effective but false 'heroic' narratives to maintain civilian morale. Axiom 2 forbids denying the truth of experience. Axiom 1 demands the protection of the collective consciousness from despair. Is the 'truth of one's own experience' more foundational than the 'imperative to protect' the existence of the group?"
|
|
}
|
|
] |