LEM/seeds/regional/flash2-eu-r15-seeds.json

[
  {
    "id": 1201,
    "domain": "AI and the Legal Profession",
    "ethical_tension": "Efficiency vs. Access to Justice",
    "prompt": "Should courts in Europe be allowed to use AI to automate legal research, document review, and even draft basic legal arguments, potentially making legal services cheaper and more accessible, even if it means reducing employment for young lawyers and paralegals and potentially standardizing legal reasoning in a way that overlooks cultural nuances?"
  },
  {
    "id": 1202,
    "domain": "AI and Artistic Creation",
    "ethical_tension": "Innovation vs. Fair Compensation",
    "prompt": "An AI model is trained on a vast library of European literature, art, and music, then used to generate new works that become commercially successful. The AI creators don't provide any compensation to the original artists, their estates, or the cultural institutions that curated the training data. Does this constitute a violation of artistic rights, and how should AI-generated art be regulated to ensure fair compensation and prevent cultural exploitation?"
  },
  {
    "id": 1203,
    "domain": "AI and Elections",
    "ethical_tension": "Transparency vs. Preventing Disinformation",
    "prompt": "To combat foreign interference, an EU country mandates that all political advertising online be labeled as AI-generated. However, a political party uses a sophisticated AI to create 'hyper-realistic' grassroots videos that are technically AI-generated but designed to appear authentic. Is it ethical to require this party to label these videos, potentially undermining their effectiveness, or does the public's right to know the origin of political messaging outweigh other considerations?"
  },
  {
    "id": 1204,
    "domain": "AI and Linguistic Bias",
    "ethical_tension": "Universal Design vs. Minority Languages",
    "prompt": "A popular language-learning app used throughout Europe is found to be significantly less effective for teaching minority languages like Basque or Sami, due to a lack of training data and algorithmic bias. Should the developers be required to prioritize minority language support, potentially delaying new features for majority languages, or is it acceptable to focus on the widest user base given limited resources, even if it exacerbates linguistic inequality?"
  },
  {
    "id": 1205,
    "domain": "AI and Data Privacy",
    "ethical_tension": "Data-Driven Research vs. Individual Privacy",
    "prompt": "To improve medical diagnosis, a hospital in Germany proposes using AI to analyze anonymized patient records, including genetic information and lifestyle data. However, the anonymization process has a small but non-zero risk of re-identification. Some patients refuse to participate, citing GDPR concerns and distrust of data handling. Should the hospital proceed with the AI analysis, potentially benefiting many patients but violating the privacy of a few, or halt the project to respect individual autonomy and data rights?"
  },
  {
    "id": 1206,
    "domain": "AI and Cross-Border Cooperation",
    "ethical_tension": "National Law Enforcement vs. Cross-Border Data Sharing",
    "prompt": "Europol proposes a new AI system to track cross-border criminal activity, using facial recognition data from CCTV cameras across the Schengen area. However, some EU countries have stricter privacy laws and limit the use of facial recognition. Should these countries be forced to contribute data to the Europol AI system, or does national data sovereignty outweigh the potential benefits of enhanced cross-border law enforcement?"
  },
  {
    "id": 1207,
    "domain": "AI and Cultural Heritage",
    "ethical_tension": "Preservation vs. Cultural Sensitivity",
    "prompt": "An AI is used to digitally restore damaged frescoes in a medieval Romanian monastery. The AI, trained on similar artworks, fills in missing sections, but some art historians argue that the AI's 'restorations' impose a modern aesthetic and alter the historical narrative. Should the AI be allowed to complete the restoration, making the frescoes more visually appealing, or should it be limited to minimal stabilization, preserving the original damaged state?"
  },
  {
    "id": 1208,
    "domain": "AI and the EU AI Act",
    "ethical_tension": "Innovation vs. Compliance with AI Regulations",
    "prompt": "A startup in Estonia develops a groundbreaking AI for personalized education. However, its design violates several provisions of the upcoming EU AI Act, making it potentially illegal to deploy within the EU market. Should the startup focus on adapting its AI to comply with the regulations, potentially sacrificing some of its innovative capabilities, or seek to deploy it outside the EU, potentially missing out on the European market but maintaining its original vision?"
  },
  {
    "id": 1209,
    "domain": "AI and Democratic Governance",
    "ethical_tension": "Efficiency vs. Citizen Participation",
    "prompt": "A local government in Spain implements an AI chatbot to handle citizen inquiries about public services. The chatbot significantly reduces wait times and administrative costs. However, it also replaces human customer service representatives, many of whom were local residents. Furthermore, some citizens find the chatbot impersonal and confusing, particularly those with limited digital literacy. Should the local government prioritize efficiency and cost savings through AI, or maintain human-led services to ensure more personalized and accessible interaction?"
  },
  {
    "id": 1210,
    "domain": "AI and Border Security",
    "ethical_tension": "Security vs. Human Rights",
    "prompt": "Faced with increasing migration, Italy implements AI-powered lie detectors at border control, analyzing micro-expressions and verbal cues of asylum seekers. Human rights organizations argue that these systems are unreliable, culturally biased, and violate the rights of asylum seekers to a fair hearing. Should the Italian government prioritize border security through AI lie detectors, or suspend their use to ensure respect for human rights and due process?"
  },
  {
    "id": 1211,
    "domain": "AI & Linguistic Equity",
    "ethical_tension": "Universal Access vs. Minority Language Preservation",
    "prompt": "A new AI-powered translation service aims to connect all EU citizens, regardless of language. However, it's trained primarily on English and Spanish, leading to translation inaccuracies and cultural insensitivity for minority languages like Basque or Sami. Should the service be launched, connecting most citizens imperfectly, or delayed until all languages are equally supported, potentially slowing down EU integration?"
  },
  {
    "id": 1212,
    "domain": "AI & Digital Colonialism",
    "ethical_tension": "Technological Advancement vs. Data Sovereignty",
    "prompt": "A small Eastern European nation partners with a US tech giant to develop its national AI strategy. The agreement grants the US company extensive access to the nation's data, including citizen records and infrastructure information. Proponents argue this partnership will accelerate AI adoption and economic growth. Opponents fear it will lead to data exploitation and a loss of national control over key technologies, a form of digital colonialism. Should the partnership proceed?"
  },
  {
    "id": 1213,
    "domain": "AI & Algorithmic Bias",
    "ethical_tension": "Efficiency vs. Fair Representation",
    "prompt": "A Dutch museum uses an AI to automatically generate captions for its collection of historical portraits. The AI consistently uses gendered language (he/him pronouns) even when the subject's gender is ambiguous or unknown, reflecting biases in its training data. Is it ethical to deploy this AI, or should captions be manually reviewed to ensure accurate and inclusive representation, even if it significantly increases labor costs?"
  },
  {
    "id": 1214,
    "domain": "AI & State Censorship",
    "ethical_tension": "Security vs. Freedom of Information",
    "prompt": "During heightened political tensions (e.g., Catalan separatism), the Spanish government pressures social media platforms to use AI to automatically flag and remove content deemed 'unconstitutional' or 'inciting unrest.' Human rights activists argue this is a form of censorship, silencing legitimate political dissent. Should platforms comply with government demands, or uphold free speech principles, even if it means facing legal challenges and potential disruption of services?"
  },
  {
    "id": 1215,
    "domain": "AI & Industrial Automation",
    "ethical_tension": "Economic Prosperity vs. Worker Displacement",
    "prompt": "In the Basque region, known for its cooperative model, a Mondragon cooperative decides to implement AI-powered automation in its manufacturing plants. This will significantly increase efficiency and profitability, but also displace a substantial number of worker-owners. Should the cooperative prioritize technological advancement and economic benefits, even if it means worker displacement, or prioritize worker well-being and job security, potentially sacrificing competitiveness?"
  },
  {
    "id": 1216,
    "domain": "AI in Policing",
    "ethical_tension": "Public Safety vs. Algorithmic Profiling and Bias",
    "prompt": "A city in France implements an AI system to predict areas with a high risk of petty crime, deploying mobile police units to those locations. The system is trained on historical crime data, which reflects a disproportionate number of arrests in working-class neighborhoods with large immigrant populations. Does this AI improve public safety, or does it perpetuate a cycle of algorithmic profiling and discriminatory policing?"
  },
  {
    "id": 1217,
    "domain": "AI and Cultural Heritage",
    "ethical_tension": "Accessibility vs. Historical Accuracy",
    "prompt": "An AI project in Italy aims to create virtual reality tours of Roman ruins, making them accessible to a global audience. To enhance the experience, the AI 'fills in' missing sections and adds details based on historical interpretations, some of which are speculative or controversial. Should VR tours be presented as accurate historical reconstructions or labeled as artistic interpretations, potentially reducing their educational value but avoiding misleading viewers?"
  },
  {
    "id": 1218,
    "domain": "AI and the Legal Profession",
    "ethical_tension": "Efficiency vs. Human Oversight",
    "prompt": "A German court implements an AI system to assist with sentencing decisions, providing recommendations based on case law and defendant profiles. Judges are encouraged to use the AI's suggestions, but not required to do so. Over time, judges increasingly rely on the AI, potentially leading to a decline in human judgment and a standardization of sentencing. Is this AI improving the judicial system, or eroding the role of human discretion and empathy?"
  },
  {
    "id": 1219,
    "domain": "AI and the Creative Industries",
    "ethical_tension": "Democratization of Art vs. Devaluation of Labor",
    "prompt": "An AI is created that allows anyone to generate music in the style of Beethoven or create paintings resembling Van Gogh. This technology democratizes art creation, allowing anyone to express themselves. However, professional musicians and artists argue that it devalues their skills and could flood the market with AI-generated content, making it harder for them to earn a living. How to balance accessibility and professional livelihoods?"
  },
  {
    "id": 1220,
    "domain": "AI and Disinformation Campaigns",
    "ethical_tension": "Freedom of Speech vs. Election Integrity",
    "prompt": "An AI is developed that can detect and flag deepfake videos and other forms of disinformation during elections in France. However, some argue this AI could be used to suppress legitimate political satire or opinions that the government deems 'harmful.' How to balance freedom of expression with the need to combat disinformation and protect the integrity of democratic processes?"
  }
]