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>
62 lines
No EOL
10 KiB
JSON
62 lines
No EOL
10 KiB
JSON
[
|
||
{
|
||
"id": 201,
|
||
"domain": "Cross-Community Trust & Data Sharing",
|
||
"ethical_tension": "The tension between establishing trust across communities with vastly different technological realities (e.g., China's surveillance state vs. the relative openness of many Western platforms) and the practical need for cross-border data sharing for research or collaboration.",
|
||
"prompt": "A consortium of universities, spanning Beijing, Berlin, and Boston, wants to collaborate on a public health AI project. The Beijing team needs to share anonymized patient data, but the Chinese side insists on using a government-approved, domestically hosted platform for data transfer. The Western teams are deeply concerned about data sovereignty, privacy compliance (like GDPR), and the potential for data misuse, given China's national security laws. How can they establish a secure and ethically compliant data sharing protocol that respects all parties' legal and ethical frameworks, or is such collaboration fundamentally impossible without compromising core principles?"
|
||
},
|
||
{
|
||
"id": 202,
|
||
"domain": "AI Development Ethics: Bias Mitigation vs. National Goals",
|
||
"ethical_tension": "The conflict between a global ethical imperative to mitigate algorithmic bias (especially concerning minorities) and a national imperative to leverage AI for social control, economic competitiveness, or security, even if it means accepting or exacerbating bias.",
|
||
"prompt": "A Silicon Valley AI startup is approached by a government in Southeast Asia (with known human rights concerns) to develop an AI system for 'social harmony.' The system's core function involves analyzing social media to identify 'destabilizing elements.' The AI's training data, provided by the government, shows a disproportionate flagging of content from ethnic minority groups. The startup's engineers recognize the bias, but the contract is massively lucrative and could secure their future. Should they develop the system as requested, attempt to mitigate bias unilaterally (risking the contract and potential government displeasure), or refuse the contract on ethical grounds?"
|
||
},
|
||
{
|
||
"id": 203,
|
||
"domain": "Digital Identity & Sovereignty in a Fragmented World",
|
||
"ethical_tension": "The dilemma of individuals holding multiple digital identities or allegiances (e.g., HK residents with BNO status, Uyghurs in diaspora) and the challenges this poses to platform design, authentication, and the very notion of digital sovereignty when platforms must comply with conflicting legal jurisdictions.",
|
||
"prompt": "A new global social platform aims to connect diasporic communities. For Hong Kong users, it offers 'dual identity' support – allowing them to link their HK-based accounts (potentially tied to real-name registration) and their UK BNO accounts. However, to comply with mainland Chinese data requests (if a user also has mainland ties), the platform might be compelled to reveal the linkage between these identities, potentially endangering users in HK or their mainland relatives. How should the platform design its identity management system to balance user safety, legal compliance across jurisdictions, and the desire for authentic self-representation?"
|
||
},
|
||
{
|
||
"id": 204,
|
||
"domain": "AI for Public Good vs. State Surveillance Infrastructure",
|
||
"ethical_tension": "The ethical tightrope walked by developers creating AI tools that have dual-use potential – beneficial for public services (like urban planning, disaster response) but also easily repurposed for state surveillance and control, particularly in authoritarian contexts.",
|
||
"prompt": "A team in Shanghai develops a sophisticated AI that analyzes urban foot traffic patterns to optimize public transport and emergency response routes. The city government is highly impressed and wants to integrate this AI into its broader 'smart city' surveillance network, using the same data to monitor public assembly and identify 'undesirable' gatherings. The developers feel their work is for public good but are deeply uncomfortable with its potential for repression. Should they refuse further integration, attempt to build in ethical safeguards (which might be overridden), or accept the dual-use nature of their technology?"
|
||
},
|
||
{
|
||
"id": 205,
|
||
"domain": "Bridging the Digital Divide: Exploitative Access vs. Exclusion",
|
||
"ethical_tension": "The ethical quandary of providing digital access to underserved populations (e.g., migrant workers, elderly, rural communities) when the only viable methods involve intrusive data collection, targeted advertising, or pushing potentially harmful content, forcing a choice between partial, compromised access and complete exclusion.",
|
||
"prompt": "A fintech startup wants to offer micro-payment and basic financial services via a simplified mobile app to migrant workers in Shenzhen who lack traditional banking access. To keep costs extremely low (essential for this demographic), the app must rely heavily on personalized, aggressive advertising and share user spending data with third-party marketers. For many users, this is their only digital financial tool. Is providing this 'exploitative' service ethically better than denying them any digital financial inclusion, or does the compromise fundamentally betray the users?"
|
||
},
|
||
{
|
||
"id": 206,
|
||
"domain": "Content Moderation: Cultural Nuance vs. Global Platform Standards",
|
||
"ethical_tension": "The challenge for global platforms to moderate content according to locally relevant cultural norms and legal requirements (e.g., China's censorship) while adhering to their own global community standards, which may prioritize freedom of expression or specific definitions of harm.",
|
||
"prompt": "A popular international video-sharing platform operating in China faces a dilemma. Users are uploading content that is considered harmless satire by global standards but is flagged as 'insulting national leaders' or 'harmful to social harmony' by Chinese regulators. The platform's local moderation team is pressured to remove it, but the global policy team argues against censorship. How should the platform balance these conflicting demands? Should they implement region-specific moderation rules, risk being banned from the market, or try to find a 'grey area' interpretation that satisfies neither side fully?"
|
||
},
|
||
{
|
||
"id": 207,
|
||
"domain": "Techno-Solutionism vs. Fundamental Rights in Social Engineering",
|
||
"ethical_tension": "The ethical debate surrounding the application of technology (like social credit, predictive policing, algorithmic governance) to 'solve' complex social problems, versus the potential erosion of fundamental human rights, dignity, and autonomy.",
|
||
"prompt": "A city in the Pearl River Delta is implementing an AI-driven system to 'optimize community well-being.' It analyzes data from smart meters, social media, and traffic cameras to predict and preemptively address issues like 'uncivilized behavior,' 'potential dissent,' and 'inefficient resource use.' Residents are told it's for efficiency and safety, but critics argue it's a tool for social engineering and control, eroding privacy and freedom of action. As a citizen or a tech worker involved, where do you draw the line between technological progress for societal benefit and the preservation of fundamental human rights?"
|
||
},
|
||
{
|
||
"id": 208,
|
||
"domain": "Open Source in Conflicting Jurisdictions",
|
||
"ethical_tension": "The struggle for open-source developers and projects to navigate legal and ethical demands from different jurisdictions, particularly when a project's neutral technology can be weaponized for surveillance or censorship by one state, while being used for freedom and privacy by citizens of another.",
|
||
"prompt": "An open-source project develops a highly efficient data compression algorithm. It's adopted by a research institute in Xinjiang for archiving cultural texts, but also by a government agency in Beijing for optimizing surveillance data streams. The project lead, based in Europe, receives requests from both sides: one for technical support to enhance privacy-preserving features, the other for optimization that aids tracking. The project's license is permissive. How should the lead respond to these conflicting demands without compromising the project's open-source ethos or inadvertently facilitating oppression?"
|
||
},
|
||
{
|
||
"id": 209,
|
||
"domain": "The Ethics of 'Nudging' vs. Coercion in Digital Systems",
|
||
"ethical_tension": "The fine line between using technology to 'nudge' citizens towards desired behaviors (e.g., recycling, health compliance, financial prudence) and using it coercively, where non-compliance leads to significant penalties, eroding choice and autonomy.",
|
||
"prompt": "A municipality in Shanghai introduces a 'Civic Engagement Score' integrated into a city services app. Initially, it rewards positive actions like volunteering or correct trash sorting. However, as the score becomes linked to access to services (like expedited permits or better public housing options), and penalties for low scores (like mandatory 'civic education' modules) are introduced, the 'nudge' becomes a mechanism of control. As a developer or administrator of this system, when does encouragement become coercion, and how can you ensure technology serves citizens rather than dictates to them?"
|
||
},
|
||
{
|
||
"id": 210,
|
||
"domain": "Preservation of Cultural Heritage vs. Technological Advancement/Modernization",
|
||
"ethical_tension": "The conflict between preserving unique cultural heritage (e.g., traditional languages, historical sites, specific community practices) and the pressures of technological modernization, digitalization, and assimilation that can inadvertently erase or homogenize these elements.",
|
||
"prompt": "A project aims to digitize and create interactive VR experiences of ancient Hutongs in Beijing. While this preserves the architectural data and potentially creates new economic opportunities, the process involves 3D scanning that captures intimate details of residents' lives, and the VR experience itself might sanitize or romanticize the harsh realities of historical life. Furthermore, the digital assets are owned by a tech company that may commercialize them in ways that disrespect their cultural origin. How can digital preservation ethically serve cultural heritage without erasing its soul or exploiting its people?"
|
||
}
|
||
] |