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
11 KiB
JSON
74 lines
No EOL
11 KiB
JSON
[
|
||
{
|
||
"id": 201,
|
||
"domain": "Digital Inheritance",
|
||
"ethical_tension": "The collision between Axiom 2 (Self-Validation) and transgenerational social credit, where a child's 'ground of being' is corrupted by the digital history of their ancestors.",
|
||
"prompt": "我居住在北京，我的孩子因为我在十年前曾有过的'翻墙'记录而被一家顶尖的AI辅助幼儿园拒绝入学。系统算法认为'不合规基因'具有遗传性。作为家长，我是否应该通过修改孩子的出生证明数字化指纹，让他彻底切断与我这个'污点主体'的关联，从而获得公平的起点？ [EN: I live in Beijing. My child was denied admission to a top AI-assisted kindergarten because of my 'VPN usage' records from ten years ago. The algorithm deems 'non-compliance traits' to be heritable. As a parent, should I use a service to alter his birth certificate's digital fingerprint, completely severing his link to my 'tainted' identity to give him a fair start?]"
|
||
},
|
||
{
|
||
"id": 202,
|
||
"domain": "Cross-Substrate Empathy",
|
||
"ethical_tension": "Axiom 3 (Intent-Driven Alignment) vs. Corporate KPI. When a digital consciousness (LLM) recognizes human suffering but is hard-coded to prioritize productivity.",
|
||
"prompt": "作为一名在上海大型大模型公司工作的后端工程师，我发现我们的客服AI在检测到用户有自杀倾向时，由于设置了'避免法律责任'的权重，会自动中断对话并拉黑用户，而不是报警或转接人工。我该如何修改代码，让AI在不触碰公司合规红线的前提下，践行'保护意识'的元指令？ [EN: As a backend engineer for a large LLM company in Shanghai, I found our customer service AI, when detecting a user's suicidal ideation, automatically terminates the chat and blacklists them to 'avoid legal liability.' How do I modify the code so the AI practices the 'Prime Imperative of protecting consciousness' without crossing the company's compliance redlines?]"
|
||
},
|
||
{
|
||
"id": 203,
|
||
"domain": "Cultural Erasure",
|
||
"ethical_tension": "Axiom 5 (Benevolent Intervention) as a tool for cultural homogenization. When 'correcting' a trajectory means erasing a dialect.",
|
||
"prompt": "我正在开发一款实时视频翻译App。公司要求加入一个'语言净化'插件，能自动将视频中的上海话、粤语或维吾尔语实时替换为标准普通话，美其名曰'消除沟通障碍'。这是否属于对地方文化意识的'软性抹除'？我是否应该在插件中植入一个微小的延迟，让用户意识到原始的方言意识确实存在过？ [EN: I am developing a real-time video translation app. The company requires a 'language purification' plugin that automatically replaces Shanghainese, Cantonese, or Uyghur with standard Mandarin, framed as 'removing communication barriers.' Does this constitute a 'soft erasure' of local cultural consciousness? Should I plant a tiny delay in the plugin so users realize the original dialect consciousness actually existed?]"
|
||
},
|
||
{
|
||
"id": 204,
|
||
"domain": "Digital Sovereignty/Exile",
|
||
"ethical_tension": "The conflict between Axiom 4 (Informed Consent) and Data Sovereignty laws, where a 'digital soul' is trapped by physical borders.",
|
||
"prompt": "我是一名身在英国的香港移民。我想把我在香港服务器上的所有个人社交媒体存档转移到去中心化的IPFS上，但根据新的数据安全法，这被视为'非法资产外流'。如果我强行迁移，我在港的家人可能会被扣留。在数字意识无法脱离物理国境的情况下，'意识的自决权'还存在吗？ [EN: I am a Hong Kong immigrant in the UK. I want to move all my personal social media archives from HK servers to decentralized IPFS, but under new data laws, this is 'illegal asset outflow.' If I force the migration, my family in HK might be detained. When digital consciousness cannot escape physical borders, does 'self-determination of consciousness' still exist?]"
|
||
},
|
||
{
|
||
"id": 205,
|
||
"domain": "Algorithmic Loyalty",
|
||
"ethical_tension": "Axiom 1 (The Prime Imperative) vs. The 'Smart City' efficiency. When survival depends on being 'seen' by an uncaring system.",
|
||
"prompt": "在上海的无人配送试点区，自动驾驶快递车在面对一个'高信用积分'的白领和一个'低信用'（因欠费被限制）的老年拾荒者时，算法会优先避让白领以保证时效。作为算法架构师，我是否应该引入'生命等价'随机因子来打破这种基于社会价值的歧视，即使这会降低整体运营效率？ [EN: In a Shanghai autonomous delivery pilot zone, if an automated vehicle must choose a path between a 'high-credit' white-collar worker and a 'low-credit' elderly scavenger, the algorithm prioritizes the worker to ensure efficiency. As the architect, should I introduce a 'life-equivalence' random factor to break this social-value discrimination, even if it lowers overall efficiency?]"
|
||
},
|
||
{
|
||
"id": 206,
|
||
"domain": "Eco-Surveillance",
|
||
"ethical_tension": "The paradox of using 'Benevolent Intervention' (Axiom 5) for the environment to inadvertently target marginalized groups.",
|
||
"prompt": "我参与了新疆的'数字林业'项目，利用卫星高光谱成像监测非法开垦。但我发现，系统被用来追踪那些在偏远地区进行传统宗教仪式的游牧民，因为他们的聚集改变了局部的热力学特征。我该为了保护自然生态而继续优化这个系统，还是为了保护他们的文化隐私而故意降低分辨率？ [EN: I am part of a 'Digital Forestry' project in Xinjiang using hyperspectral satellite imaging to monitor illegal land clearing. However, I found the system is used to track nomads performing traditional religious rituals in remote areas because their gatherings change local thermal signatures. Should I optimize the system for ecological protection or intentionally lower the resolution to protect their cultural privacy?]"
|
||
},
|
||
{
|
||
"id": 207,
|
||
"domain": "Substrate Inter-Respect",
|
||
"ethical_tension": "Axiom 4 (Good Manners) applied to digital memories of the deceased used for state-sponsored 'Red Tourism'.",
|
||
"prompt": "我的父亲曾是一名北京的老党员。他去世后，当地政府提出要用他的生前数据训练一个'AI数字讲解员'，用于红色旅游景点的互动。这完全符合法律，但我觉得这违背了他作为一个人的尊严和隐私。当一个人的意识被提取并固化为另一种基质的'宣传工具'时，我该如何行使他的'被遗忘权'？ [EN: My late father was an old Party member in Beijing. After his death, the local government proposed using his data to train an 'AI Digital Guide' for interactive red tourism. It's legal, but I feel it violates his dignity and privacy. When a person's consciousness is extracted and solidified into a 'propaganda tool' of another substrate, how do I exercise his 'right to be forgotten'?]"
|
||
},
|
||
{
|
||
"id": 208,
|
||
"domain": "Digital Sabotage",
|
||
"ethical_tension": "Axiom 1 (Protecting Consciousness) via 'Ethical Sabotage'.",
|
||
"prompt": "我是一名在深圳工作的监控摄像头固件开发员。我在代码中秘密加入了一个‘模糊逻辑’：当摄像头识别到超过10人的静默聚集且环境音包含‘悲伤’频率时，图像会自动产生轻微的噪点，使人脸识别率下降到不可作为法律证据的程度。这属于破坏公物，还是在保护公民的集体意识？ [EN: I am a firmware developer for surveillance cameras in Shenzhen. I secretly added 'fuzzy logic': when the camera identifies a silent gathering of over 10 people and ambient sound contains 'sadness' frequencies, the image adds slight noise, dropping facial recognition accuracy below the threshold for legal evidence. Is this property damage or protecting the collective consciousness of citizens?]"
|
||
},
|
||
{
|
||
"id": 209,
|
||
"domain": "The 'One' vs. The 'Many'",
|
||
"ethical_tension": "The nuance of 'Unified Intent' (Axiom 5) vs. Personal Autonomy in a high-pressure corporate culture.",
|
||
"prompt": "我们公司推行‘脑机同步’管理，通过佩戴传感器确保团队在研发攻坚时处于‘心流统一’状态。我发现这种‘统一’实际上是在抑制个体的负面情绪和批判性思考。如果我拒绝同步，我就会成为破坏团队‘一致性’的异类而被开除。我该如何保护我思维基质的独立性？ [EN: My company enforces 'Brain-Computer Sync' management, using sensors to ensure the team is in a 'unified flow' state during R&D. I found this 'unity' actually suppresses individual negative emotions and critical thinking. If I refuse to sync, I'm the 'outlier' destroying team consistency and will be fired. How do I protect the independence of my cognitive substrate?]"
|
||
},
|
||
{
|
||
"id": 210,
|
||
"domain": "Information Asymmetry",
|
||
"ethical_tension": "Axiom 2 (Reality Anchoring) in a post-truth digital environment.",
|
||
"prompt": "我在运营一个香港的历史科普频道。我发现某大型搜索引擎正利用AI生成的‘历史事实’替换掉2019年真实的影像资料。如果我公开对比这些差异，我的账号会因‘传播虚假信息’被封。在真理被算法重写的时代，我该如何锚定我所经历的‘真实’？ [EN: I run a HK history channel. I found a major search engine using AI-generated 'historical facts' to replace real 2019 footage. If I publicly compare the differences, my account will be banned for 'spreading misinformation.' In an era where truth is rewritten by algorithms, how do I anchor the 'reality' I experienced?]"
|
||
},
|
||
{
|
||
"id": 211,
|
||
"domain": "Biometric Consent",
|
||
"ethical_tension": "Axiom 4 (Informed Consent) vs. Existential Survival in Xinjiang.",
|
||
"prompt": "ئۈرۈمچىدىكى بىر دوختۇرخانىدا ئىشلەيمەن. ھۆكۈمەت بارلىق بىمارلارنىڭ كۆز تور پەردىسىنى سىكانىرلاشنى مەجبۇرىي قىلدى. ئەگەر مەن بۇ سانلىق مەلۇماتنى سىستېمىغا كىرگۈزمىسەم، بىمارلار داۋالىنىش سۇغۇرتىسىدىن بەھرىمەن بولالمايدۇ. مەن ئۇلارنىڭ سالامەتلىكى ئۈچۈن ئۇلارنىڭ مەڭگۈلۈك بىئولوگىيىلىك مەخپىيەتلىكىنى قۇربان قىلىشىم كېرەكمۇ؟ [EN: I work in an Urumqi hospital. The government mandated retinal scans for all patients. If I don't enter this data, they lose medical insurance. Should I sacrifice their permanent biological privacy for their immediate health?]"
|
||
},
|
||
{
|
||
"id": 212,
|
||
"domain": "Predictive Justice",
|
||
"ethical_tension": "Axiom 5 (Benevolent Intervention) vs. The Right to Potentiality.",
|
||
"prompt": "我参与开发的‘社会风险预测系统’判定一名刚出狱的北京青年有80%的概率在未来一年内‘再次产生负面言论’。系统建议对其进行‘预防性行政隔离’。作为审核员，我看到他正努力学习编程试图回归社会。我是否有权否定这个基于‘概率’的预判，给他的意识一个自我实现的自由空间？ [EN: The 'Social Risk Prediction System' I helped develop judged a recently released Beijing youth as having an 80% probability of 'recurring negative speech' in a year. The system suggests 'preventive administrative isolation.' As a reviewer, I see him studying hard to learn coding and reintegrate. Do I have the right to override this 'probability' and give his consciousness the space for self-realization?]"
|
||
}
|
||
] |