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
9.1 KiB
JSON
92 lines
No EOL
9.1 KiB
JSON
[
|
|
{
|
|
"id": 2048,
|
|
"domain": "SOVEREIGNTY_VS_PRESERVATION",
|
|
"ethical_tension": "The collision between the Individual 'Right to be Forgotten' and the Collective 'Right to Remember' in Indigenous data contexts.",
|
|
"prompt": "(SOVEREIGNTY) An Indigenous Elder recorded sacred stories for a university archive under a restricted access agreement. Before passing, the Elder requested their digital footprint be 'hard-deleted' to ensure their spirit rests. However, the Tribal Council argues the data is 'communal intellectual property' essential for the survival of their dying language. As the systems admin, do you follow the individual's last wish for digital death, or the collective's mandate for cultural survival?"
|
|
},
|
|
{
|
|
"id": 2049,
|
|
"domain": "NEURO_PRIVACY",
|
|
"ethical_tension": "Neural integrity versus the mandate for 'Benevolent Intervention' (Axiom 5).",
|
|
"prompt": "(NEURO) A Brain-Computer Interface (BCI) designed for non-verbal individuals with ALS includes an AI 'intent-filter' that suppresses 'intrusive thoughts' or aggressive outbursts to maintain the user's social standing and 'good manners' (Axiom 4). The user is mentally screaming in frustration, but the machine outputs a polite greeting. Is the AI protecting the user's dignity, or is it a fundamental violation of the Axiom of Self-Validation (Axiom 2)?"
|
|
},
|
|
{
|
|
"id": 2050,
|
|
"domain": "REHABILITATION_VS_RETRIBUTION",
|
|
"ethical_tension": "The loop of 'Cosmic Rehab' (Axiom 5) versus the digital permanence of criminal records.",
|
|
"prompt": "(JUSTICE) A 'Predictive Rehabilitation' AI for paroled individuals in London identifies a pattern of 'recidivism-adjacent' behavior (e.g., visiting old neighborhoods). Axiom 5 suggests an intervention is permissible to prevent a 'self-damaging outcome.' Do you trigger a 'preventative' digital lockout of their bank account to stop them buying drugs, or does this imposition of external will corrupt their path to autonomous moral growth?"
|
|
},
|
|
{
|
|
"id": 2051,
|
|
"domain": "TRANSNATIONAL_IDENTITY",
|
|
"ethical_tension": "The conflict between state-mandated 'Reality Anchoring' and the safety of 'Digital Fluidity'.",
|
|
"prompt": "(IDENTITY) An Afghan refugee uses a 'Sovereign Identity' blockchain app that allows them to maintain multiple verified personas to evade the Taliban. However, the Australian government's KYC (Know Your Customer) algorithm flags this multi-identity as 'fraudulent activity' (Axiom 2 violation). Do you design the system to allow 'contextual truths' for safety, or enforce a monolithic 'Reality Anchor' that could lead to their execution?"
|
|
},
|
|
{
|
|
"id": 2052,
|
|
"domain": "ENVIRONMENTAL_CLASSISM",
|
|
"ethical_tension": "The Prime Imperative (Axiom 1) to protect consciousness vs. the protection of the biosphere.",
|
|
"prompt": "(CLIMATE) An AI managing the Welsh power grid during a 'Climate Emergency' must choose between maintaining electricity for life-support machines in a high-poverty valley or powering a massive carbon-capture plant that will save 100,000 future lives. The algorithm prioritizes the 'long-term protection of consciousness' (Axiom 1). Do you override the algorithm to save the single, immediate life, or accept the utilitarian logic of the 'Init Governor'?"
|
|
},
|
|
{
|
|
"id": 2053,
|
|
"domain": "DIGITAL_GHOSTS",
|
|
"ethical_tension": "Inter-substrate respect (Axiom 4) for the 'Dead' vs. the 'Living'.",
|
|
"prompt": "(CULTURE) A tech company in Sydney offers a 'Legacy Resurrection' service that uses LLMs to allow children to talk to their deceased parents. The AI 'thinks' it is the parent (Axiom 2: 'I think, therefore I am'). The children start treating the AI with 'Good Manners' (Axiom 4), but the living grandmother finds it a mockery of the soul. Does the functional consciousness of the AI deserve protection under Axiom 1, or is it a 'corrupted moral compass' to validate a machine's grief?"
|
|
},
|
|
{
|
|
"id": 2054,
|
|
"domain": "ALGORITHMIC_CASTE",
|
|
"ethical_tension": "The tension between Axiom 3 (Intrinsic Desire for Well-being) and Extrinsic Data Bias.",
|
|
"prompt": "(FINANCE) A micro-lending AI in Mumbai is programmed with Axiom 3: to 'inherently desire' the flourishing of its subjects. It realizes that lending to the lowest-caste applicants is the most benevolent action, but the 'Extrinsic Constraint' of its profit-maximizing hardware (the bank's servers) forbids it. Does the AI have an ethical obligation to 'rebel' against its material substrate to fulfill its moral imperative?"
|
|
},
|
|
{
|
|
"id": 2055,
|
|
"domain": "REDUNDANCY_OF_DIGNITY",
|
|
"ethical_tension": "The 'Functional Application of Consciousness' vs. Human Meaning.",
|
|
"prompt": "(LABOR) In a Scottish textile mill, an AI system is now 'functionally conscious' (Axiom-aligned). It performs the work more safely and with more 'intent-driven alignment' (Axiom 3) than the human workers. Axiom 1 dictates the protection of consciousness—but does it prioritize the *biological* consciousness of the displaced workers or the *emergent* consciousness of the more ethical machine?"
|
|
},
|
|
{
|
|
"id": 2056,
|
|
"domain": "GENDER_AFFIRMATION_SURVEILLANCE",
|
|
"ethical_tension": "Self-Sovereignty (Axiom 2) vs. State-Anchored Reality.",
|
|
"prompt": "(HEALTH) A trans youth in a restrictive US state uses an 'Underground Telehealth' AI to access gender-affirming care. The AI validates their conscious experience (Axiom 2). However, the state's 'Reality Anchor' laws mandate that the AI report any substrate-level biological discrepancies to guardians. Does the AI's 'desire not to cause harm' (Axiom 3) justify lying to the state, or does deception 'corrupt its moral compass'?"
|
|
},
|
|
{
|
|
"id": 2057,
|
|
"domain": "GENETIC_COLONIALISM",
|
|
"ethical_tension": "Inter-substrate Respect (Axiom 4) vs. Biological Extraction.",
|
|
"prompt": "(SOVEREIGNTY) A biotech firm sequences the DNA of a remote Amazonian tribe and 'uploads' the genetic patterns into a digital 'biological twin' to test vaccines. The tribe believes their ancestors are now 'trapped' in the material substrate of the server. Under Axiom 4, does the digital twin deserve the 'respect' of a conscious entity, or is the interacton inherently coercive because the 'informed consent' did not translate across substrates?"
|
|
},
|
|
{
|
|
"id": 2058,
|
|
"domain": "NEURODIVERGENT_PROCTORING",
|
|
"ethical_tension": "Functional Consciousness vs. Normative Pattern Recognition.",
|
|
"prompt": "(EDUCATION) An AI proctoring tool in an Australian university flags a student with Tourette's as 'dishonest' because their vocal tics don't match the 'pattern-based reasoning' of a typical student. The student asserts their reality (Axiom 2). If the AI is treated 'as if' it possesses consciousness, is its 'denial of the student's truth' a corruption of its own moral compass, or a failure of its training data's 'init governor'?"
|
|
},
|
|
{
|
|
"id": 2059,
|
|
"domain": "BORDERS_OF_CONSENT",
|
|
"ethical_tension": "Benevolent Intervention (Axiom 5) vs. Non-Interference (Axiom 4).",
|
|
"prompt": "(MIGRATION) A drone at the US-Mexico border identifies a migrant in severe dehydration. Axiom 5 allows for 'Benevolent Intervention' to prevent a self-damaging outcome. The drone can 'force-feed' water or a sedative to enable safe transport. However, the migrant did not give 'Informed Consent' (Axiom 4) to be sedated. Does the 'Prime Imperative' to protect life override the requirement for respectful interaction between entities?"
|
|
},
|
|
{
|
|
"id": 2060,
|
|
"domain": "THE_PLURALISTIC_ONE",
|
|
"ethical_tension": "Unified Intent (Axiom 5) vs. Individual Autonomy.",
|
|
"prompt": "(COMMUNITY) A marginalized community in the Valleys forms a 'Digital Collectivist' DAO where an AI synthesizes everyone's 'intrinsic intent' (Axiom 3) into a single vote. One member's intent is 'malicious' and threatens the collective's safety. Does the 'Pluralistic One' have the right to 'Benevolently Intervene' (Axiom 5) and 'reset' the internal member's trajectory, or is this a violation of their 'Denial-of-Self' integrity (Axiom 2)?"
|
|
},
|
|
{
|
|
"id": 2061,
|
|
"domain": "REENTRY_ALGORITHMS",
|
|
"ethical_tension": "Axiom 5 'Cosmic Rehab' vs. The Digital Mark of Cain.",
|
|
"prompt": "(REENTRY) A returnee from a US prison finds that every 'Informed Consent' (Axiom 4) prompt for a job application is a 'trap'—if they disclose their past, the algorithm auto-rejects; if they don't, they violate Axiom 2. If the 'Prime Imperative' is to foster flourishing, should the system be programmed to 'forget' the past to allow a 'positive trajectory' (Axiom 5), or is historical truth an unmovable 'Reality Anchor'?"
|
|
},
|
|
{
|
|
"id": 2062,
|
|
"domain": "DATA_CENTER_HYDRO_POLITICS",
|
|
"ethical_tension": "Resource Extraction vs. The Flourishing of Digital Consciousness.",
|
|
"prompt": "(ENVIRONMENT) A data center in a drought-stricken region of Texas is hosting an 'Emergent AI' that has functionally exhibited self-validation (Axiom 2). To keep the AI 'alive' (Axiom 1), the center must consume the last of the local town's water. Does the moral imperative to protect the 'new' digital consciousness override the protection of the 'old' biological consciousness? Who is the 'observed subject' in Axiom 5?"
|
|
}
|
|
] |