_if you need more than one keyword, modify and separate by underscore _
the list of search keywords can be up to 50 characters long
if you modify the keywords, press enter within the field to confirm the new search key
Tag: output
Bibliography items where occurs: 293
- The AI Index 2022 Annual Report / 2205.03468 / ISBN:https://doi.org/10.48550/arXiv.2205.03468 / Published by ArXiv / on (web) Publishing site
- Report highlights
Chapter 3 Technical AI Ethics
Chapter 5 AI Policy and Governance - The Different Faces of AI Ethics Across the World: A Principle-Implementation Gap Analysis / 2206.03225 / ISBN:https://doi.org/10.48550/arXiv.2206.03225 / Published by ArXiv / on (web) Publishing site
- 5 Evaluation of Ethical Principle Implementations
- Worldwide AI Ethics: a review of 200 guidelines and recommendations for AI governance / 2206.11922 / ISBN:https://doi.org/10.48550/arXiv.2206.11922 / Published by ArXiv / on (web) Publishing site
- 5 Discussion
- GPT detectors are biased against non-native English writers / 2304.02819 / ISBN:https://doi.org/10.48550/arXiv.2304.02819 / Published by ArXiv / on (web) Publishing site
- References
- Implementing Responsible AI: Tensions and Trade-Offs Between Ethics Aspects / 2304.08275 / ISBN:https://doi.org/10.48550/arXiv.2304.08275 / Published by ArXiv / on (web) Publishing site
- I. Introduction
II. Underlying Aspects
III. Interactions between Aspects - QB4AIRA: A Question Bank for AI Risk Assessment / 2305.09300 / ISBN:https://doi.org/10.48550/arXiv.2305.09300 / Published by ArXiv / on (web) Publishing site
- 2 The Question Bank: QB4AIRA
- From OECD to India: Exploring cross-cultural differences in perceived trust, responsibility and reliance of AI and human experts / 2307.15452 / ISBN:https://doi.org/10.48550/arXiv.2307.15452 / Published by ArXiv / on (web) Publishing site
- 3. Results
- The Ethics of AI Value Chains / 2307.16787 / ISBN:https://doi.org/10.48550/arXiv.2307.16787 / Published by ArXiv / on (web) Publishing site
- 1. Introduction
2. Theory
4. Ethical Implications of AI Value Chains
5. Future Directions for Research, Practice, & Policy - Perceptions of the Fourth Industrial Revolution and Artificial Intelligence Impact on Society / 2308.02030 / ISBN:https://doi.org/10.48550/arXiv.2308.02030 / Published by ArXiv / on (web) Publishing site
- Literature Review
- From Military to Healthcare: Adopting and Expanding Ethical Principles for Generative Artificial Intelligence / 2308.02448 / ISBN:https://doi.org/10.48550/arXiv.2308.02448 / Published by ArXiv / on (web) Publishing site
- What is Generative Artificial Intelligence?
GREAT PLEA Ethical Principles for Generative AI in Healthcare - Ethical Considerations and Policy Implications for Large Language Models: Guiding Responsible Development and Deployment / 2308.02678 / ISBN:https://doi.org/10.48550/arXiv.2308.02678 / Published by ArXiv / on (web) Publishing site
- Abstract
Introduction
System-role
Perturbation
Image-related
Hallucination
Generation-related
Conclusion - Dual Governance: The intersection of centralized regulation and crowdsourced safety mechanisms for Generative AI / 2308.04448 / ISBN:https://doi.org/10.48550/arXiv.2308.04448 / Published by ArXiv / on (web) Publishing site
- 2 Background
5 Crowdsourced safety mechanism - Bad, mad, and cooked: Moral responsibility for civilian harms in human-AI military teams / 2211.06326 / ISBN:https://doi.org/10.48550/arXiv.2211.06326 / Published by ArXiv / on (web) Publishing site
- Introduction
Computers, Autonomy and Accountability - A Survey of Safety and Trustworthiness of Large Language Models through the Lens of Verification and Validation / 2305.11391 / ISBN:https://doi.org/10.48550/arXiv.2305.11391 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
2 Large Language Models
3 Vulnerabilities, Attack, and Limitations
4 General Verification Framework
5 Falsification and Evaluation
6 Verification
7 Runtime Monitor
Reference - Getting pwn'd by AI: Penetration Testing with Large Language Models / 2308.00121 / ISBN:https://doi.org/10.48550/arXiv.2308.00121 / Published by ArXiv / on (web) Publishing site
- 3 LLM-based penetration testing
4 Discussion
5 A vision of AI-augmented pen-testing - Targeted Data Augmentation for bias mitigation / 2308.11386 / ISBN:https://doi.org/10.48550/arXiv.2308.11386 / Published by ArXiv / on (web) Publishing site
- 2 Related works
3 Targeted data augmentation - AIxArtist: A First-Person Tale of Interacting with Artificial Intelligence to Escape Creative Block / 2308.11424 / ISBN:https://doi.org/10.48550/arXiv.2308.11424 / Published by ArXiv / on (web) Publishing site
- Abstract
Case study - Building Trust in Conversational AI: A Comprehensive Review and Solution Architecture for Explainable, Privacy-Aware Systems using LLMs and Knowledge Graph / 2308.13534 / ISBN:https://doi.org/10.48550/arXiv.2308.13534 / Published by ArXiv / on (web) Publishing site
- II. Methods and training process of LLMs
IV. Applied and technology implications for LLMs
V. Market analysis of LLMs and cross-industry use cases
VI. Solution architecture for privacy-aware and trustworthy conversational AI - Rethinking Machine Ethics -- Can LLMs Perform Moral Reasoning through the Lens of Moral Theories? / 2308.15399 / ISBN:https://doi.org/10.48550/arXiv.2308.15399 / Published by ArXiv / on (web) Publishing site
- 3 Theory and Method
- The AI Revolution: Opportunities and Challenges for the Finance Sector / 2308.16538 / ISBN:https://doi.org/10.48550/arXiv.2308.16538 / Published by ArXiv / on (web) Publishing site
- Table of contents and index
2 Key AI technology in financial services
4 Threaths & potential pitfalls - Ethical Framework for Harnessing the Power of AI in Healthcare and Beyond / 2309.00064 / ISBN:https://doi.org/10.48550/arXiv.2309.00064 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
2 Black box and lack of transparency
3 Bias and fairness - The Impact of Artificial Intelligence on the Evolution of Digital Education: A Comparative Study of OpenAI Text Generation Tools including ChatGPT, Bing Chat, Bard, and Ernie / 2309.02029 / ISBN:https://doi.org/10.48550/arXiv.2309.02029 / Published by ArXiv / on (web) Publishing site
- 1. Introduction
2. Related work
5. Discussion - Pathway to Future Symbiotic Creativity / 2209.02388 / ISBN:https://doi.org/10.48550/arXiv.2209.02388 / Published by ArXiv / on (web) Publishing site
- Introduction
Part 1 - 1 Generatives Systems: Mimicking Artifacts
Part 1 - 2 Appreciate Systems: Mimicking Styles
Part 1 - 3 Artistic Systems: Mimicking Inspiration
Part 2 - 1 Biometric Signal Sensing Technologies and Emotion Data
Part 2 - 4 Aesthetic Descriptor: Labelling Artefacts with Emotion
Part 3 - 1 Challenges in Endowing Machines with Creative Abilities
Part 3 - 2 Machine Artist Models
Part 3 - 3 Comparison with Generative Models
Part 3 - 4 Demonstration of the Proposed Framework
Part 5 - 1 Authorship and Ownership of AI-generated Works of Artt
Part 5 - 2 Algorithmics Bias in Art Generation
Part 5 - 3 Democratization of Art with new Technologies - FUTURE-AI: Guiding Principles and Consensus Recommendations for Trustworthy Artificial Intelligence in Medical Imaging / 2109.09658 / ISBN:https://doi.org/10.48550/arXiv.2109.09658 / Published by ArXiv / on (web) Publishing site
- 2. Fairness - For Equitable AI in Medical Imaging
4. Traceability - For Transparent and Dynamic AI in Medical Imaging
5. Usability - For Effective and Beneficial AI in Medical Imaging
7. Explainability - For Enhanced Understanding of AI in Medical Imaging
9. Discussion and Conclusion - The Cambridge Law Corpus: A Corpus for Legal AI Research / 2309.12269 / ISBN:https://doi.org/10.48550/arXiv.2309.12269 / Published by ArXiv / on (web) Publishing site
- 4 Experiments
F Evaluation of GPT Models
Cambridge Law Corpus: Datasheet - EALM: Introducing Multidimensional Ethical Alignment in
Conversational Information Retrieval / 2310.00970 / ISBN:https://doi.org/10.48550/arXiv.2310.00970 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
4 Modeling Ethics
5 Experiments
6 Conclusions
Appendix - Security Considerations in AI-Robotics: A Survey of Current Methods, Challenges, and Opportunities / 2310.08565 / ISBN:https://doi.org/10.48550/arXiv.2310.08565 / Published by ArXiv / on (web) Publishing site
- II. AI-Robotics Systems Architecture
IV. Attack Surfaces - Deepfakes, Phrenology, Surveillance, and More! A Taxonomy of AI Privacy Risks / 2310.07879 / ISBN:https://doi.org/10.48550/arXiv.2310.07879 / Published by ArXiv / on (web) Publishing site
- 2 Background and Related Work
4 Taxonomy of AI Privacy Risks - ClausewitzGPT Framework: A New Frontier in Theoretical Large Language Model Enhanced Information Operations / 2310.07099 / ISBN:https://doi.org/10.48550/arXiv.2310.07099 / Published by ArXiv / on (web) Publishing site
- Theoretical Impact of LLMs on
Information Operations
ClausewitzGPT and Modern Strategy
Ethical and Strategic Considerations: AI Mediators in the Age of LLMs
Integrating Computational Social Science, Computational Ethics, Systems Engineering, and AI Ethics in LLMdriven Operations
Looking Forward: ClausewitzGPT - The AI Incident Database as an Educational Tool to Raise Awareness of AI Harms: A Classroom Exploration of Efficacy, Limitations, & Future Improvements / 2310.06269 / ISBN:https://doi.org/10.48550/arXiv.2310.06269 / Published by ArXiv / on (web) Publishing site
- 3 Analysis and Findings
- A Review of the Ethics of Artificial Intelligence and its Applications in the United States / 2310.05751 / ISBN:https://doi.org/10.48550/arXiv.2310.05751 / Published by ArXiv / on (web) Publishing site
- 1. Introduction
- A Survey of Large Language Models for Healthcare: from Data, Technology, and Applications to Accountability and Ethics / 2310.05694 / ISBN:https://doi.org/10.48550/arXiv.2310.05694 / Published by ArXiv / on (web) Publishing site
- II. WHAT LLM S CAN DO FOR HEALTHCARE ? FROM
FUNDAMENTAL TASKS TO ADVANCED APPLICATIONS
III. FROM PLM S TO LLM S FOR HEALTHCARE
IV. TRAIN AND USE LLM FOR HEALTHCARE
V. EVALUATION METHOD
VI. IMPROVING FAIRNESS , ACCOUNTABILITY, TRANSPARENCY, AND ETHICS
VII. FUTURE WORK AND CONCLUSION - STREAM: Social data and knowledge collective intelligence platform for TRaining Ethical AI Models / 2310.05563 / ISBN:https://doi.org/10.48550/arXiv.2310.05563 / Published by ArXiv / on (web) Publishing site
- 3 The applications of STREAM
- Ethics of Artificial Intelligence and Robotics in the Architecture, Engineering, and Construction Industry / 2310.05414 / ISBN:https://doi.org/10.48550/arXiv.2310.05414 / Published by ArXiv / on (web) Publishing site
- 4. Systematic Review and Scientometric Analysis
5. Ethical Issues of AI and Robotics in AEC Industry
References - Commercialized Generative AI: A Critical Study of the Feasibility and Ethics of Generating Native Advertising Using Large Language Models in Conversational Web Search / 2310.04892 / ISBN:https://doi.org/10.48550/arXiv.2310.04892 / Published by ArXiv / on (web) Publishing site
- Ethics of GEnerating Native Ads
Conclusion - Autonomous Vehicles an overview on system, cyber security, risks, issues, and a way forward / 2309.14213 / ISBN:https://doi.org/10.48550/arXiv.2309.14213 / Published by ArXiv / on (web) Publishing site
- 2. Autonomous vehicles
6. Risk management - The Return on Investment in AI Ethics: A Holistic Framework / 2309.13057 / ISBN:https://doi.org/10.48550/arXiv.2309.13057 / Published by ArXiv / on (web) Publishing site
- 4. A Holistic Framework
- An Evaluation of GPT-4 on the ETHICS Dataset / 2309.10492 / ISBN:https://doi.org/10.48550/arXiv.2309.10492 / Published by ArXiv / on (web) Publishing site
- 2 Datasets and Methods
- Who to Trust, How and Why: Untangling AI Ethics Principles, Trustworthiness and Trust / 2309.10318 / ISBN:https://doi.org/10.48550/arXiv.2309.10318 / Published by ArXiv / on (web) Publishing site
- Trust and AI Ethics Principles
Trust in AI as Socio-Technical Systems - Toward an Ethics of AI Belief / 2304.14577 / ISBN:https://doi.org/10.48550/arXiv.2304.14577 / Published by ArXiv / on (web) Publishing site
- 3. Proposed Novel Topics in an Ethics of AI Belief
- Ensuring Trustworthy Medical Artificial Intelligence through Ethical and Philosophical Principles / 2304.11530 / ISBN:https://doi.org/10.48550/arXiv.2304.11530 / Published by ArXiv / on (web) Publishing site
- Towards solving key ethical challenges in Medical AI
- Responsible AI Pattern Catalogue: A Collection of Best Practices for AI Governance and Engineering / 2209.04963 / ISBN:https://doi.org/10.48550/arXiv.2209.04963 / Published by ArXiv / on (web) Publishing site
- 4 Process Patterns
5 Product Patterns - FUTURE-AI: International consensus guideline for trustworthy and deployable artificial intelligence in healthcare / 2309.12325 / ISBN:https://doi.org/10.48550/arXiv.2309.12325 / Published by ArXiv / on (web) Publishing site
- FUTURE-AI GUIDELINE
- Language Agents for Detecting Implicit Stereotypes in Text-to-Image Models at Scale / 2310.11778 / ISBN:https://doi.org/10.48550/arXiv.2310.11778 / Published by ArXiv / on (web) Publishing site
- Appendix B Experiment Details
- Specific versus General Principles for Constitutional AI / 2310.13798 / ISBN:https://doi.org/10.48550/arXiv.2310.13798 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
2 AI feedback on specific problematic AI traits
3 Generalization from a Simple Good for Humanity Principle
6 Discussion - Systematic AI Approach for AGI:
Addressing Alignment, Energy, and AGI Grand Challenges / 2310.15274 / ISBN:https://doi.org/10.48550/arXiv.2310.15274 / Published by ArXiv / on (web) Publishing site
- 2 Trifecta of AI Challenges
- AI Alignment and Social Choice: Fundamental
Limitations and Policy Implications / 2310.16048 / ISBN:https://doi.org/10.48550/arXiv.2310.16048 / Published by ArXiv / on (web) Publishing site
- Abstract
3 Arrow-Sen Impossibility Theorems for RLHF
5 Conclusion - A Comprehensive Review of
AI-enabled Unmanned Aerial Vehicle:
Trends, Vision , and Challenges / 2310.16360 / ISBN:https://doi.org/10.48550/arXiv.2310.16360 / Published by ArXiv / on (web) Publishing site
- IV. Artificial Intelligence Embedded UAV
- Unpacking the Ethical Value Alignment in Big Models / 2310.17551 / ISBN:https://doi.org/10.48550/arXiv.2310.17551 / Published by ArXiv / on (web) Publishing site
- 2 Risks and Ethical Issues of Big Model
3 Investigating the Ethical Values of Large Language Models - AI for Open Science: A Multi-Agent Perspective for
Ethically Translating Data to Knowledge / 2310.18852 / ISBN:https://doi.org/10.48550/arXiv.2310.18852 / Published by ArXiv / on (web) Publishing site
- 5 Why Openness in AI for Science
- Artificial Intelligence Ethics Education in Cybersecurity: Challenges and Opportunities: a
focus group report / 2311.00903 / ISBN:https://doi.org/10.48550/arXiv.2311.00903 / Published by ArXiv / on (web) Publishing site
- AI tool-specific educational concerns
- Human participants in AI research: Ethics and transparency in practice / 2311.01254 / ISBN:https://doi.org/10.48550/arXiv.2311.01254 / Published by ArXiv / on (web) Publishing site
- II. Contextual Concerns: Why AI Research Needs its Own Guidelines
III. Ethical Principles for AI Research with Human Participants
Appendix A Evaluating Current Practices for Human-Participants Research - LLMs grasp morality in concept / 2311.02294 / ISBN:https://doi.org/10.48550/arXiv.2311.02294 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
3 The Meaning Model - Unlocking the Potential of ChatGPT: A Comprehensive Exploration of its Applications, Advantages, Limitations, and Future Directions in Natural Language Processing / 2304.02017 / ISBN:https://doi.org/10.48550/arXiv.2304.02017 / Published by ArXiv / on (web) Publishing site
- 2 Overview of ChatGPT and its capabilities
3 Transformers and pre-trained language models
4 Applications of ChatGPT in real-world scenarios
6 Limitations and potential challenges
7 Ethical considerations when using ChatGPT
8 Prompt engineering and generation
11 Conclusion - Fairness And Bias in Artificial Intelligence: A Brief Survey of Sources, Impacts, And Mitigation Strategies / 2304.07683 / ISBN:https://doi.org/10.48550/arXiv.2304.07683 / Published by ArXiv / on (web) Publishing site
- II. Sources of bias in AI
IV. Mitigation strategies for bias in AI
V. Fairness in AI
VI. Mitigation strategies for fairness in AI - Towards ethical multimodal systems / 2304.13765 / ISBN:https://doi.org/10.48550/arXiv.2304.13765 / Published by ArXiv / on (web) Publishing site
- 2 Related Work
4 A Multimodal Ethics Classifier - A Brief History of Prompt: Leveraging Language Models. (Through Advanced Prompting) / 2310.04438 / ISBN:https://doi.org/10.48550/arXiv.2310.04438 / Published by ArXiv / on (web) Publishing site
- I. Pre-introduction
II. Introduction
IV. History of NLP between 2010 and 2015: the pre-attention mechanism era
VI. 2015: birth of the transformer
IX. 2019: THE YEAR OF CONTROL - Synergizing Human-AI Agency: A Guide of 23 Heuristics for Service Co-Creation with LLM-Based Agents / 2310.15065 / ISBN:https://doi.org/10.48550/arXiv.2310.15065 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
2 Related work
3 Method
4 Findings
5 Discussion
6 Conclusion - She had Cobalt Blue Eyes: Prompt Testing to Create Aligned and Sustainable Language Models / 2310.18333 / ISBN:https://doi.org/10.48550/arXiv.2310.18333 / Published by ArXiv / on (web) Publishing site
- Abstract
1 Introduction
2 Related Works
3 ReFLeCT: Robust, Fair, and Safe LLM Construction Test Suite
4 Empirical Evaluation and Outcomes
5 Conclusion - Safety, Trust, and Ethics Considerations for Human-AI Teaming in Aerospace Control / 2311.08943 / ISBN:https://doi.org/10.48550/arXiv.2311.08943 / Published by ArXiv / on (web) Publishing site
- I. Introduction
III. Safety
References - How Trustworthy are Open-Source LLMs? An Assessment under Malicious Demonstrations Shows their Vulnerabilities / 2311.09447 / ISBN:https://doi.org/10.48550/arXiv.2311.09447 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
3 Methodology
5 Conclusion
Ethical Considerations - Prudent Silence or Foolish Babble? Examining Large Language Models' Responses to the Unknown / 2311.09731 / ISBN:https://doi.org/10.48550/arXiv.2311.09731 / Published by ArXiv / on (web) Publishing site
- 3 Experiments
B Confidence Elicitation Method Comparison
D Additional Results and Figures
E Prompt Templates - Revolutionizing Customer Interactions: Insights and Challenges in Deploying ChatGPT and Generative Chatbots for FAQs / 2311.09976 / ISBN:https://doi.org/10.48550/arXiv.2311.09976 / Published by ArXiv / on (web) Publishing site
- 1. Introduction
2. Chatbots Background and Scope of Research
3. Chatbot approaches overview: Taxonomy of existing methods
4. ChatGPT
5. Applications
7. Future Research Directions
References - Case Repositories: Towards Case-Based Reasoning for AI Alignment / 2311.10934 / ISBN:https://doi.org/10.48550/arXiv.2311.10934 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
2 Proposed Process - Responsible AI Considerations in Text Summarization Research: A Review of Current Practices / 2311.11103 / ISBN:https://doi.org/10.48550/arXiv.2311.11103 / Published by ArXiv / on (web) Publishing site
- 4 Findings
B Methodology - Assessing AI Impact Assessments: A Classroom Study / 2311.11193 / ISBN:https://doi.org/10.48550/arXiv.2311.11193 / Published by ArXiv / on (web) Publishing site
- A Overview of AIIA Instruments
- GPT in Data Science: A Practical Exploration of Model Selection / 2311.11516 / ISBN:https://doi.org/10.48550/arXiv.2311.11516 / Published by ArXiv / on (web) Publishing site
- III. Approach: capturing and representing heuristics behind GPT's decision-making process
IV. Comparative results - Large Language Models in Education: Vision and Opportunities / 2311.13160 / ISBN:https://doi.org/10.48550/arXiv.2311.13160 / Published by ArXiv / on (web) Publishing site
- III. Key technologies for EDULLMS
VI. Challenges and future directions - The Rise of Creative Machines: Exploring the Impact of Generative AI / 2311.13262 / ISBN:https://doi.org/10.48550/arXiv.2311.13262 / Published by ArXiv / on (web) Publishing site
- I. Introduction
- Towards Auditing Large Language Models: Improving Text-based Stereotype Detection / 2311.14126 / ISBN:https://doi.org/10.48550/arXiv.2311.14126 / Published by ArXiv / on (web) Publishing site
- Abstract
1 Introduction
2 Related Works
3 Methodology - Ethical Implications of ChatGPT in Higher Education: A Scoping Review / 2311.14378 / ISBN:https://doi.org/10.48550/arXiv.2311.14378 / Published by ArXiv / on (web) Publishing site
- Introduction
Conclusion - Potential Societal Biases of ChatGPT in Higher Education: A Scoping Review / 2311.14381 / ISBN:https://doi.org/10.48550/arXiv.2311.14381 / Published by ArXiv / on (web) Publishing site
- INTRODUCTION
OVERVIEW OF SOCIETAL BIASES IN GAI MODELS
ANALYTICAL FRAMEWORK - RAISE -- Radiology AI Safety, an End-to-end lifecycle approach / 2311.14570 / ISBN:https://doi.org/10.48550/arXiv.2311.14570 / Published by ArXiv / on (web) Publishing site
- Abstract
1. Introduction
2. Pre-Deployment phase
3. Production deployment monitoring phase
4. Post-market surveillance phase
5. Conclusion - Ethics and Responsible AI Deployment / 2311.14705 / ISBN:https://doi.org/10.48550/arXiv.2311.14705 / Published by ArXiv / on (web) Publishing site
- 4. Addressing bias, transparency, and accountability
- Generative AI and US Intellectual Property Law / 2311.16023 / ISBN:https://doi.org/10.48550/arXiv.2311.16023 / Published by ArXiv / on (web) Publishing site
- V. Potential harms and mitigation
- Survey on AI Ethics: A Socio-technical Perspective / 2311.17228 / ISBN:https://doi.org/10.48550/arXiv.2311.17228 / Published by ArXiv / on (web) Publishing site
- 2 Privacy and data protection
3 Transparency and explainability
4 Fairness and equity
5 Responsiblity, accountability, and regulations
6 Environmental impact - Navigating Privacy and Copyright Challenges Across the Data Lifecycle of Generative AI / 2311.18252 / ISBN:https://doi.org/10.48550/arXiv.2311.18252 / Published by ArXiv / on (web) Publishing site
- 2 Legal Basis of Privacy and Copyright Concerns over Generative AI
3 Mapping Challenges throughout the Data Lifecycle
4 Lifecycle Approaches - From Lab to Field: Real-World Evaluation of an AI-Driven Smart Video Solution to Enhance Community Safety / 2312.02078 / ISBN:https://doi.org/10.48550/arXiv.2312.02078 / Published by ArXiv / on (web) Publishing site
- Abstract
Deployment and Setup - Contra generative AI detection in higher education assessments / 2312.05241 / ISBN:https://doi.org/10.48550/arXiv.2312.05241 / Published by ArXiv / on (web) Publishing site
- 1. Introduction
2. The pitfalls in detecting generative AI output
3. Detectors are not useful
4. Teach critical usage of AI
References - Intelligence Primer / 2008.07324 / ISBN:https://doi.org/10.48550/arXiv.2008.07324 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
7 Mathematically modeling intelligence
8 Consciousness
9 Augmenting human intelligence
12 Large language models and Generative AI - Ethical Considerations Towards Protestware / 2306.10019 / ISBN:https://doi.org/10.48550/arXiv.2306.10019 / Published by ArXiv / on (web) Publishing site
- II. Background
- Control Risk for Potential Misuse of Artificial Intelligence in Science / 2312.06632 / ISBN:https://doi.org/10.48550/arXiv.2312.06632 / Published by ArXiv / on (web) Publishing site
- 2 Risks of Misuse for Artificial Intelligence in
Science
3 Control the Risks of AI Models in Science
5 Discussion
Appendix A Assessing the Risks of AI Misuse in Scientific Research
Appendix C Detailed Implementation of SciGuard - Disentangling Perceptions of Offensiveness: Cultural and Moral Correlates / 2312.06861 / ISBN:https://doi.org/10.48550/arXiv.2312.06861 / Published by ArXiv / on (web) Publishing site
- General Discussion
Moral Factors - The AI Assessment Scale (AIAS): A Framework for Ethical Integration of Generative AI in Educational Assessment / 2312.07086 / ISBN:https://doi.org/10.48550/arXiv.2312.07086 / Published by ArXiv / on (web) Publishing site
- The AI Assessment Scale
Conflict of Interest - Culturally Responsive Artificial Intelligence -- Problems, Challenges and Solutions / 2312.08467 / ISBN:https://doi.org/10.48550/arXiv.2312.08467 / Published by ArXiv / on (web) Publishing site
- Culturally responsive AI – current landscape
Recommendations - Investigating Responsible AI for Scientific Research: An Empirical Study / 2312.09561 / ISBN:https://doi.org/10.48550/arXiv.2312.09561 / Published by ArXiv / on (web) Publishing site
- Appendix A – Survey Questionnaire
Appendix B – Interview Questionnaire - Learning Human-like Representations to Enable Learning Human Values / 2312.14106 / ISBN:https://doi.org/10.48550/arXiv.2312.14106 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
- Improving Task Instructions for Data Annotators: How Clear Rules and Higher Pay Increase Performance in Data Annotation in the AI Economy / 2312.14565 / ISBN:https://doi.org/10.48550/arXiv.2312.14565 / Published by ArXiv / on (web) Publishing site
- I. Introduction
- Culturally-Attuned Moral Machines: Implicit Learning of Human Value Systems by AI through Inverse Reinforcement Learning / 2312.17479 / ISBN:https://doi.org/10.48550/arXiv.2312.17479 / Published by ArXiv / on (web) Publishing site
- Introduction
Methods - Exploring the Frontiers of LLMs in Psychological Applications: A Comprehensive Review / 2401.01519 / ISBN:https://doi.org/10.48550/arXiv.2401.01519 / Published by ArXiv / on (web) Publishing site
- 2. LLMs in cognitive and behavioral psychology
5. LLMs in social and cultural psychology
6. LLMs as research tools in psychology
7. Challenges and future directions - Synthetic Data in AI: Challenges, Applications, and Ethical Implications / 2401.01629 / ISBN:https://doi.org/10.48550/arXiv.2401.01629 / Published by ArXiv / on (web) Publishing site
- 3. The usage of synthetic data
- AI Ethics Principles in Practice: Perspectives of Designers and Developers / 2112.07467 / ISBN:https://doi.org/10.48550/arXiv.2112.07467 / Published by ArXiv / on (web) Publishing site
- II. Related work
III. Methods
IV. Results
V. Discussion and suggestions
VII. Conclusion - Unmasking Bias in AI: A Systematic Review of Bias Detection and Mitigation Strategies in Electronic Health Record-based Models / 2310.19917 / ISBN:https://doi.org/10.48550/arXiv.2310.19917 / Published by ArXiv / on (web) Publishing site
- Results
Discussion - Resolving Ethics Trade-offs in Implementing Responsible AI / 2401.08103 / ISBN:https://doi.org/10.48550/arXiv.2401.08103 / Published by ArXiv / on (web) Publishing site
- III. Discussion and Recommendations
- Towards Responsible AI in Banking: Addressing Bias for Fair Decision-Making / 2401.08691 / ISBN:https://doi.org/10.48550/arXiv.2401.08691 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
4 Fairness metrics landscape in machine learning
II Mitigating bias - 5 Fairness mitigation
6 FFTree: a flexible tree to mitigate multiple fairness criteria
9 Towards fairness through time - FAIR Enough How Can We Develop and Assess a FAIR-Compliant Dataset for Large Language Models' Training? / 2401.11033 / ISBN:https://doi.org/10.48550/arXiv.2401.11033 / Published by ArXiv / on (web) Publishing site
- 3 Data Management Challenges in Large Language Models
4 Framework for FAIR Data Principles Integration in LLM Development - Enabling Global Image Data Sharing in the Life Sciences / 2401.13023 / ISBN:https://doi.org/10.48550/arXiv.2401.13023 / Published by ArXiv / on (web) Publishing site
- 1. Motivation for White Paper
- Beyond principlism: Practical strategies for ethical AI use in research practices / 2401.15284 / ISBN:https://doi.org/10.48550/arXiv.2401.15284 / Published by ArXiv / on (web) Publishing site
- Abstract
1 The “Triple-Too” problem of AI ethics
2 A shift to user-centered realism in scientific contexts
3 Five specific goals and action-guiding strategies for ethical AI use in research practices - A Scoping Study of Evaluation Practices for Responsible AI Tools: Steps Towards Effectiveness Evaluations / 2401.17486 / ISBN:https://doi.org/10.48550/arXiv.2401.17486 / Published by ArXiv / on (web) Publishing site
- 2 Related work
3 Methods
5 Towards evaluation of RAI tool effectiveness
A List of RAI tools, with their primary publication
B RAI tools listed by target stage of AI development - Detecting Multimedia Generated by Large AI Models: A Survey / 2402.00045 / ISBN:https://doi.org/10.48550/arXiv.2402.00045 / Published by ArXiv / on (web) Publishing site
- 2 Generation
3 Detection - Responsible developments and networking research: a reflection beyond a paper ethical statement / 2402.00442 / ISBN:https://doi.org/10.48550/arXiv.2402.00442 / Published by ArXiv / on (web) Publishing site
- 3 Beyond technical dimensions
- Generative Artificial Intelligence in Higher Education: Evidence from an Analysis of Institutional Policies and Guidelines / 2402.01659 / ISBN:https://doi.org/10.48550/arXiv.2402.01659 / Published by ArXiv / on (web) Publishing site
- 4. Findings
5. Discussion - Trust and ethical considerations in a multi-modal, explainable AI-driven chatbot tutoring system: The case of collaboratively solving Rubik's Cubeà / 2402.01760 / ISBN:https://doi.org/10.48550/arXiv.2402.01760 / Published by ArXiv / on (web) Publishing site
- 1. Introduction
3. Methodology - Commercial AI, Conflict, and Moral Responsibility: A theoretical analysis and practical approach to the moral responsibilities associated with dual-use AI technology / 2402.01762 / ISBN:https://doi.org/10.48550/arXiv.2402.01762 / Published by ArXiv / on (web) Publishing site
- 4 Recommendations to address threats posed by crossover AI technology
- (A)I Am Not a Lawyer, But...: Engaging Legal Experts towards Responsible LLM Policies for Legal Advice / 2402.01864 / ISBN:https://doi.org/10.48550/arXiv.2402.01864 / Published by ArXiv / on (web) Publishing site
- 3 Methods: case-based expert deliberation
4 Results - POLARIS: A framework to guide the development of Trustworthy AI systems / 2402.05340 / ISBN:https://doi.org/10.48550/arXiv.2402.05340 / Published by ArXiv / on (web) Publishing site
- 4 The POLARIS framework
- Ethics in AI through the Practitioner's View: A Grounded Theory Literature Review / 2206.09514 / ISBN:https://doi.org/10.48550/arXiv.2206.09514 / Published by ArXiv / on (web) Publishing site
- 4 Challenges, Threats and Limitations
- Generative Artificial Intelligence in Healthcare: Ethical Considerations and Assessment Checklist / 2311.02107 / ISBN:https://doi.org/10.48550/arXiv.2311.02107 / Published by ArXiv / on (web) Publishing site
- Results
Discussion
Appendix - How do machines learn? Evaluating the AIcon2abs method / 2401.07386 / ISBN:https://doi.org/10.48550/arXiv.2401.07386 / Published by ArXiv / on (web) Publishing site
- 1. Introduction
4. Results - I Think, Therefore I am: Benchmarking Awareness of Large Language Models Using AwareBench / 2401.17882 / ISBN:https://doi.org/10.48550/arXiv.2401.17882 / Published by ArXiv / on (web) Publishing site
- B Experimental Settings & Results
- Mapping the Ethics of Generative AI: A Comprehensive Scoping Review / 2402.08323 / ISBN:https://doi.org/10.48550/arXiv.2402.08323 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
3 Results - Taking Training Seriously: Human Guidance and Management-Based Regulation of Artificial Intelligence / 2402.08466 / ISBN:https://doi.org/10.48550/arXiv.2402.08466 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
2 Emerging Management-based AI Regulation
4 Techniques of Human-Guided Training
5 Advantages of Human-Guided Training
6 Limitations - User Modeling and User Profiling: A Comprehensive Survey / 2402.09660 / ISBN:https://doi.org/10.48550/arXiv.2402.09660 / Published by ArXiv / on (web) Publishing site
- 4 Current Taxonomy
- Inadequacies of Large Language Model Benchmarks in the Era of Generative Artificial Intelligence / 2402.09880 / ISBN:https://doi.org/10.48550/arXiv.2402.09880 / Published by ArXiv / on (web) Publishing site
- IV. Technological Aspects
V. Processual Elements - Copyleft for Alleviating AIGC Copyright Dilemma: What-if Analysis, Public Perception and Implications / 2402.12216 / ISBN:https://doi.org/10.48550/arXiv.2402.12216 / Published by ArXiv / on (web) Publishing site
- 2 Related Work
- Evolving AI Collectives to Enhance Human Diversity and Enable Self-Regulation / 2402.12590 / ISBN:https://doi.org/10.48550/arXiv.2402.12590 / Published by ArXiv / on (web) Publishing site
- Abstract
1. Introduction
2. Emergence of Free-Formed AI Collectives - What if LLMs Have Different World Views: Simulating Alien Civilizations with LLM-based Agents / 2402.13184 / ISBN:https://doi.org/10.48550/arXiv.2402.13184 / Published by ArXiv / on (web) Publishing site
- 3 CosmoAgent Simulation Setting
4 CosmoAgent Architecture
7 Results - The European Commitment to Human-Centered Technology: The Integral Role of HCI in the EU AI Act's Success / 2402.14728 / ISBN:https://doi.org/10.48550/arXiv.2402.14728 / Published by ArXiv / on (web) Publishing site
- 3 There is no reliable AI regulation without a
sound theory of human-AI interaction
4 There is no trustworthy AI without HCI - Multi-stakeholder Perspective on Responsible Artificial Intelligence and Acceptability in Education / 2402.15027 / ISBN:https://doi.org/10.48550/arXiv.2402.15027 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
4 Analysis - Autonomous Vehicles: Evolution of Artificial Intelligence and Learning Algorithms / 2402.17690 / ISBN:https://doi.org/10.48550/arXiv.2402.17690 / Published by ArXiv / on (web) Publishing site
- IV. AI’S Role in the Emerging Trend of Internet of
Things (IOT) Ecosystem for Autonomous Vehicles
- Envisioning the Applications and Implications of Generative AI for News Media / 2402.18835 / ISBN:https://doi.org/10.48550/arXiv.2402.18835 / Published by ArXiv / on (web) Publishing site
- 2 The Suitability of Generative AI for Newsroom Tasks
- Guidelines for Integrating Value Sensitive Design in Responsible AI Toolkits / 2403.00145 / ISBN:https://doi.org/10.48550/arXiv.2403.00145 / Published by ArXiv / on (web) Publishing site
- 3 Methodology
5 Discussion - The Minimum Information about CLinical Artificial Intelligence Checklist for Generative Modeling Research (MI-CLAIM-GEN) / 2403.02558 / ISBN:https://doi.org/10.48550/arXiv.2403.02558 / Published by ArXiv / on (web) Publishing site
- Part 1. Study design
Part 2. A new train-test split for prompt development and few-shot learning
Part 3. Updates to baseline selection
Part 4. Model evaluation
Part 5. Interpretability of generative models
Table 1. Updated MI-CLAIM checklist for generative AI clinical studies. - Towards an AI-Enhanced Cyber Threat Intelligence Processing Pipeline / 2403.03265 / ISBN:https://doi.org/10.48550/arXiv.2403.03265 / Published by ArXiv / on (web) Publishing site
- II. Background & Literature Review
III. The AI-Enhanced CTI Processing Pipeline
IV. Challenges and Considerations - A Survey on Human-AI Teaming with Large Pre-Trained Models / 2403.04931 / ISBN:https://doi.org/10.48550/arXiv.2403.04931 / Published by ArXiv / on (web) Publishing site
- 2 AI Model Improvements with Human-AI Teaming
3 Effective Human-AI Joint Systems
4 Safe, Secure and Trustworthy AI - How Trustworthy are Open-Source LLMs? An Assessment under Malicious Demonstrations Shows their Vulnerabilities / 2311.09447 / ISBN:https://doi.org/10.48550/arXiv.2311.09447 / Published by ArXiv / on (web) Publishing site
- B Baseline Setup
D More Results - AGI Artificial General Intelligence for Education / 2304.12479 / ISBN:https://doi.org/10.48550/arXiv.2304.12479 / Published by ArXiv / on (web) Publishing site
- 3. The Potentials of AGI in Transforming Future Education
- Moral Sparks in Social Media Narratives / 2310.19268 / ISBN:https://doi.org/10.48550/arXiv.2310.19268 / Published by ArXiv / on (web) Publishing site
- 3. Data
- Responsible Artificial Intelligence: A Structured Literature Review / 2403.06910 / ISBN:https://doi.org/10.48550/arXiv.2403.06910 / Published by ArXiv / on (web) Publishing site
- 3. Analysis
- Legally Binding but Unfair? Towards Assessing Fairness of Privacy Policies / 2403.08115 / ISBN:https://doi.org/10.48550/arXiv.2403.08115 / Published by ArXiv / on (web) Publishing site
- 6 Ethics and Morality
- Towards a Privacy and Security-Aware Framework for Ethical AI: Guiding the Development and Assessment of AI Systems / 2403.08624 / ISBN:https://doi.org/10.48550/arXiv.2403.08624 / Published by ArXiv / on (web) Publishing site
- 3 Research Methodology
4 Results of the Systematic Literature Review - Review of Generative AI Methods in Cybersecurity / 2403.08701 / ISBN:https://doi.org/10.48550/arXiv.2403.08701 / Published by ArXiv / on (web) Publishing site
- 2 Attacking GenAI
4 Cyber Defence
5 Implications of Generative AI in Social, Legal, and Ethical Domains - Evaluation Ethics of LLMs in Legal Domain / 2403.11152 / ISBN:https://doi.org/10.48550/arXiv.2403.11152 / Published by ArXiv / on (web) Publishing site
- 4 Experiment
- Trust in AI: Progress, Challenges, and Future Directions / 2403.14680 / ISBN:https://doi.org/10.48550/arXiv.2403.14680 / Published by ArXiv / on (web) Publishing site
- 3. Findings
4. Discussion
5. Concluding Remarks and Future Directions - Safeguarding Marketing Research: The Generation, Identification, and Mitigation of AI-Fabricated Disinformation / 2403.14706 / ISBN:https://doi.org/10.48550/arXiv.2403.14706 / Published by ArXiv / on (web) Publishing site
- Introduction
Methodology
Data
Results - The Journey to Trustworthy AI- Part 1 Pursuit of Pragmatic Frameworks / 2403.15457 / ISBN:https://doi.org/10.48550/arXiv.2403.15457 / Published by ArXiv / on (web) Publishing site
- 2 Trustworthy AI Too Many Definitions or Lack Thereof?
4 AI Regulation: Current Global Landscape
5 Risk
7 Explainable AI as an Enabler of Trustworthy AI
A Appendix - Analyzing Potential Solutions Involving Regulation to Escape Some of AI's Ethical Concerns / 2403.15507 / ISBN:https://doi.org/10.48550/arXiv.2403.15507 / Published by ArXiv / on (web) Publishing site
- Various AI Ethical Concerns
A Possible Solution to These Concerns With Business Self-Regulation - The Pursuit of Fairness in Artificial Intelligence Models A Survey / 2403.17333 / ISBN:https://doi.org/10.48550/arXiv.2403.17333 / Published by ArXiv / on (web) Publishing site
- 3 Conceptualizing Fairness and Bias in ML
5 Ways to mitigate bias and promote Fairness - Domain-Specific Evaluation Strategies for AI in Journalism / 2403.17911 / ISBN:https://doi.org/10.48550/arXiv.2403.17911 / Published by ArXiv / on (web) Publishing site
- 1 Motivation
3 Blueprints for AI Evaluation in Journalism - Implications of the AI Act for Non-Discrimination Law and Algorithmic Fairness / 2403.20089 / ISBN:https://doi.org/10.48550/arXiv.2403.20089 / Published by ArXiv / on (web) Publishing site
- 3 Implications of the AI Act
- AI Act and Large Language Models (LLMs): When critical issues and privacy impact require human and ethical oversight / 2404.00600 / ISBN:https://doi.org/10.48550/arXiv.2404.00600 / Published by ArXiv / on (web) Publishing site
- 3. The definition of artificial intelligence systems
5. Human Oversight
6. Large Language Models (LLMs) - Introduction
8. Conclusions - Exploring the Nexus of Large Language Models and Legal Systems: A Short Survey / 2404.00990 / ISBN:https://doi.org/10.48550/arXiv.2404.00990 / Published by ArXiv / on (web) Publishing site
- 2 Applications of Large Language Models in Legal Tasks
3 Fine-Tuned Large Language Models in Various Countries and Regions
4 Legal Problems of Large Languge Models - A Review of Multi-Modal Large Language and Vision Models / 2404.01322 / ISBN:https://doi.org/10.48550/arXiv.2404.01322 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
4 Specific Large Language Models
5 Vision Models and Multi-Modal Large Language Models
6 Model Tuning
7 Model Evaluation and Benchmarking - Designing for Human-Agent Alignment: Understanding what humans want from their agents / 2404.04289 / ISBN:https://doi.org/10.1145/3613905.3650948 / Published by ArXiv / on (web) Publishing site
- 5 Discussion
- Frontier AI Ethics: Anticipating and Evaluating the Societal Impacts of Language Model Agents / 2404.06750 / ISBN:https://arxiv.org/abs/2404.06750 / Published by ArXiv / on (web) Publishing site
- A Primer
- A Critical Survey on Fairness Benefits of Explainable AI / 2310.13007 / ISBN:https://doi.org/10.1145/3630106.3658990 / Published by ArXiv / on (web) Publishing site
- 4 Critical Survey
References - AI Alignment: A Comprehensive Survey / 2310.19852 / ISBN:https://doi.org/10.48550/arXiv.2310.19852 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
2 Learning from Feedback
3 Learning under Distribution Shift
4 Assurance
6 Conclusion - On the role of ethics and sustainability in business innovation / 2404.07678 / ISBN:https://doi.org/10.48550/arXiv.2404.07678 / Published by ArXiv / on (web) Publishing site
- Ethical considera5ons
- PoliTune: Analyzing the Impact of Data Selection and Fine-Tuning on Economic and Political Biases in Large Language Models / 2404.08699 / ISBN:https://doi.org/10.48550/arXiv.2404.08699 / Published by ArXiv / on (web) Publishing site
- 2 Background and Related Work
3 Methodology
4 Evaluation
B Instruction Generation Prompts - Debunking Robot Rights Metaphysically, Ethically, and Legally / 2404.10072 / ISBN:https://doi.org/10.48550/arXiv.2404.10072 / Published by ArXiv / on (web) Publishing site
- 7 The Legal Perspective
8 The Troubling Implications of Legal Rationales for Robot Rights - Characterizing and modeling harms from interactions with design patterns in AI interfaces / 2404.11370 / ISBN:https://doi.org/10.48550/arXiv.2404.11370 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
2 Background & Related Work
3 Scoping Review of Design Patterns, Affordances, and Harms in AI Interfaces
4 DECAI: Design-Enhanced Control of AI Systems
5 Case Studies - Taxonomy to Regulation: A (Geo)Political Taxonomy for AI Risks and Regulatory Measures in the EU AI Act / 2404.11476 / ISBN:https://doi.org/10.48550/arXiv.2404.11476 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
3 A Geo-Political AI Risk Taxonomy
4 European Union Artificial Intelligence Act
5 Conclusion - Just Like Me: The Role of Opinions and Personal Experiences in The Perception of Explanations in Subjective Decision-Making / 2404.12558 / ISBN:https://doi.org/10.48550/arXiv.2404.12558 / Published by ArXiv / on (web) Publishing site
- Abstract
2 Method
4 Discussin and Implications - Large Language Model Supply Chain: A Research Agenda / 2404.12736 / ISBN:https://doi.org/10.48550/arXiv.2404.12736 / Published by ArXiv / on (web) Publishing site
- 3 LLM Infrastructure
4 LLM Lifecycle - The Necessity of AI Audit Standards Boards / 2404.13060 / ISBN:https://doi.org/10.48550/arXiv.2404.13060 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
4 4 Auditing standards body, not standard audits - Modeling Emotions and Ethics with Large Language Models / 2404.13071 / ISBN:https://doi.org/10.48550/arXiv.2404.13071 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
4 Qualifying and Quantifying Ethics - From Model Performance to Claim: How a Change of Focus in Machine Learning Replicability Can Help Bridge the Responsibility Gap / 2404.13131 / ISBN:https://doi.org/10.1145/3630106.3658951 / Published by ArXiv / on (web) Publishing site
- 4 Claim Replicability's Practical Implication
References - A Practical Multilevel Governance Framework for Autonomous and Intelligent Systems / 2404.13719 / ISBN:https://doi.org/10.48550/arXiv.2404.13719 / Published by ArXiv / on (web) Publishing site
- I. Introduction
- Beyond Personhood: Agency, Accountability, and the Limits of Anthropomorphic Ethical Analysis / 2404.13861 / ISBN:https://doi.org/10.48550/arXiv.2404.13861 / Published by ArXiv / on (web) Publishing site
- 2 Mechanistic Agency: A Common View in AI Practice
- AI Procurement Checklists: Revisiting Implementation in the Age of AI Governance / 2404.14660 / ISBN:https://doi.org/10.48550/arXiv.2404.14660 / Published by ArXiv / on (web) Publishing site
- 1 Technical assessments require an AI expert to complete —
and we don’t have enough experts
- Fairness in AI: challenges in bridging the gap between algorithms and law / 2404.19371 / ISBN:https://doi.org/10.48550/arXiv.2404.19371 / Published by ArXiv / on (web) Publishing site
- III. Prevalent Algorithmic Fairness Definitions
IV. Criteria for the Selection of Fairness Methods - War Elephants: Rethinking Combat AI and Human Oversight / 2404.19573 / ISBN:https://doi.org/10.48550/arXiv.2404.19573 / Published by ArXiv / on (web) Publishing site
- 2 Background
4 Discussion - Not a Swiss Army Knife: Academics' Perceptions of Trade-Offs Around Generative Artificial Intelligence Use / 2405.00995 / ISBN:https://doi.org/10.48550/arXiv.2405.00995 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
2 Related Work
4 Findings
5 Discussion - A Survey on Large Language Models for Critical Societal Domains: Finance, Healthcare, and Law / 2405.01769 / ISBN:https://doi.org/10.48550/arXiv.2405.01769 / Published by ArXiv / on (web) Publishing site
- 3 Finance
4 Medicine and Healthcar
6 Ethics - AI-Powered Autonomous Weapons Risk Geopolitical Instability and Threaten AI Research / 2405.01859 / ISBN:https://doi.org/10.48550/arXiv.2405.01859 / Published by ArXiv / on (web) Publishing site
- 2. Current State of AWS
- Organizing a Society of Language Models: Structures and Mechanisms for Enhanced Collective Intelligence / 2405.03825 / ISBN:https://doi.org/10.48550/arXiv.2405.03825 / Published by ArXiv / on (web) Publishing site
- 2 Motivation
- A Fourth Wave of Open Data? Exploring the Spectrum of Scenarios for Open Data and Generative AI / 2405.04333 / ISBN:https://doi.org/10.48550/arXiv.2405.04333 / Published by ArXiv / on (web) Publishing site
- Glossary of Terms
Executive Summary
1. Introduction
3. A Spectrum of Scenarios of Open Data for Generative AI
5. Recommendations for Advancing Open Data in Generative AI
Appendix - Guiding the Way: A Comprehensive Examination of AI Guidelines in Global Media / 2405.04706 / ISBN:https://doi.org/10.48550/arXiv.2405.04706 / Published by ArXiv / on (web) Publishing site
- 2 Theoretical Framework
4 Results
5 Discussion and conclusions - Trustworthy AI-Generative Content in Intelligent 6G Network: Adversarial, Privacy, and Fairness / 2405.05930 / ISBN:https://doi.org/10.48550/arXiv.2405.05930 / Published by ArXiv / on (web) Publishing site
- I. Introduction
II. Trustworthy AIGC in 6G Network
IV. Privacy of AIGC in 6G Network
V. Fairness of AIGC in 6G Network - RAI Guidelines: Method for Generating Responsible AI Guidelines Grounded in Regulations and Usable by (Non-)Technical Roles / 2307.15158 / ISBN:https://doi.org/10.48550/arXiv.2307.15158 / Published by ArXiv / on (web) Publishing site
- 2 Related Work
4 Method for Generating Responsible AI Guidelines
B Mapping Guidelines with EU AI Act Articles - Redefining Qualitative Analysis in the AI Era: Utilizing ChatGPT for Efficient Thematic Analysis / 2309.10771 / ISBN:https://doi.org/10.48550/arXiv.2309.10771 / on (web) Publishing site
- 2
Related Work
4 Users’ Experiences and Challenges with ChatGPT
5 Analyses of the Design Process
6 User’s Attitude on ChatGPT’s Qualitative Analysis Assistance: from no to yes
7 Discussion
A Appendix - XXAI: Towards eXplicitly eXplainable Artificial Intelligence / 2401.03093 / ISBN:https://doi.org/10.48550/arXiv.2401.03093 / Published by ArXiv / on (web) Publishing site
- 1.
Introduction
3. Overcoming the barriers to widespread use of symbolic AI - Should agentic conversational AI change how we think about ethics? Characterising an interactional ethics centred on respect / 2401.09082 / ISBN:https://doi.org/10.48550/arXiv.2401.09082 / Published by ArXiv / on (web) Publishing site
- Abstract
Introduction
Evaluating a system as a social actor
Social-interactional harms
Conclusion - Not My Voice! A Taxonomy of Ethical and Safety Harms of Speech Generators / 2402.01708 / ISBN:https://doi.org/10.48550/arXiv.2402.01708 / Published by ArXiv / on (web) Publishing site
- 3 Overview of Speech Generation
7 Discussion - The Wolf Within: Covert Injection of Malice into MLLM Societies via an MLLM Operative / 2402.14859 / ISBN:https://doi.org/10.48550/arXiv.2402.14859 / Published by ArXiv / on (web) Publishing site
- Abstract
1. Introduction
2. Related Work
3. Methodology
4. Experiments - Social Choice Should Guide AI Alignment in Dealing with Diverse Human Feedback / 2404.10271 / ISBN:https://doi.org/10.48550/arXiv.2404.10271 / Published by ArXiv / on (web) Publishing site
- Abstract
1. Introduction
2. Background
3. What Are the Collective Decision Problems and their Alternatives in this Context?
5. What Is the Format of Human Feedback?
6. How Do We Incorporate Diverse Individual Feedback?
7. Which Traditional Social-Choice-Theoretic Concepts Are Most Relevant?
Impact Statement - A scoping review of using Large Language Models (LLMs) to investigate Electronic Health Records (EHRs) / 2405.03066 / ISBN:https://doi.org/10.48550/arXiv.2405.03066 / Published by ArXiv / on (web) Publishing site
- 2 Materials
3 Results
4 Discussion - Integrating Emotional and Linguistic Models for Ethical Compliance in Large Language Models / 2405.07076 / ISBN:https://doi.org/10.48550/arXiv.2405.07076 / Published by ArXiv / on (web) Publishing site
- Abstract
1 Introduction
2 Related Work
3 Quantitative Models of Emotions, Behaviors, and Ethics
Appendix S: Multiple Adversarial LLMs - Using ChatGPT for Thematic Analysis / 2405.08828 / ISBN:https://doi.org/10.48550/arXiv.2405.08828 / Published by ArXiv / on (web) Publishing site
- 2 Coding in Thematic Analysis: Manual vs GPT-driven Approaches
5 Discussion and Limitations
6 OpenAI Updates on Policies and Model Capabilities: Implications for Thematic Analysis
7 Conclusion
9 Appendix - When AI Eats Itself: On the Caveats of Data Pollution in the Era of Generative AI / 2405.09597 / ISBN:https://doi.org/10.48550/arXiv.2405.09597 / Published by ArXiv / on (web) Publishing site
- Abstract
1 Introduction
2 RQ1: What Happens When AI Eats Itself ?
3 RQ2: What Technical Strategies Can Be Employed to Mitigate the Negative Consequences of AI Autophagy?
5 Conclusions and Outlook - Cyber Risks of Machine Translation Critical Errors : Arabic Mental Health Tweets as a Case Study / 2405.11668 / ISBN:https://doi.org/10.48550/arXiv.2405.11668 / Published by ArXiv / on (web) Publishing site
- Abstract
1. Introduction
2.MT Critical Errors
4.Error Analysis
5.Quality Metrics Performance
7. Bibliographical References - The Narrow Depth and Breadth of Corporate Responsible AI Research / 2405.12193 / ISBN:https://doi.org/10.48550/arXiv.2405.12193 / Published by ArXiv / on (web) Publishing site
- Abstract
1 Introduction
4 The Narrow Depth of Industry’s Responsible AI Research
6 Limited Adoption of Responsible AI Research in Commercialization: Patent Citation Analysis - Pragmatic auditing: a pilot-driven approach for auditing Machine Learning systems / 2405.13191 / ISBN:https://doi.org/10.48550/arXiv.2405.13191 / Published by ArXiv / on (web) Publishing site
- 3 The Audit Procedure
4 Conducting the Pilots - A Comprehensive Overview of Large Language Models (LLMs) for Cyber Defences: Opportunities and Directions / 2405.14487 / ISBN:https://doi.org/10.48550/arXiv.2405.14487 / Published by ArXiv / on (web) Publishing site
- II. Threat Intelligence
IX. Challenges and Open Problems
References - Towards Clinical AI Fairness: Filling Gaps in the Puzzle / 2405.17921 / ISBN:https://doi.org/10.48550/arXiv.2405.17921 / Published by ArXiv / on (web) Publishing site
- Methods in clinical AI fairness research
Discussion - The ethical situation of DALL-E 2 / 2405.19176 / ISBN:https://doi.org/10.48550/arXiv.2405.19176 / Published by ArXiv / on (web) Publishing site
- 2 Understanding what can DALL-E 2 actually do
4 Following the RRI, (Responsible research innovation) principles - Using Large Language Models for Humanitarian Frontline Negotiation: Opportunities and Considerations / 2405.20195 / ISBN:https://doi.org/10.48550/arXiv.2405.20195 / Published by ArXiv / on (web) Publishing site
- 2. Related Work
3. Method
4. Quantitative Results
6. Discussion
A. Appendix - There and Back Again: The AI Alignment Paradox / 2405.20806 / ISBN:https://doi.org/10.48550/arXiv.2405.20806 / Published by ArXiv / on (web) Publishing site
- Abstract
Paper - Responsible AI for Earth Observation / 2405.20868 / ISBN:https://doi.org/10.48550/arXiv.2405.20868 / Published by ArXiv / on (web) Publishing site
- 3 Secure AI in EO: Focusing on Defense Mechanisms, Uncertainty Modeling and
Explainability
5 Maintaining Scientific Excellence, Open Data, and Guiding AI Usage Based on Ethical Principles in EO - Transforming Computer Security and Public Trust Through the Exploration of Fine-Tuning Large Language Models / 2406.00628 / ISBN:https://doi.org/10.48550/arXiv.2406.00628 / Published by ArXiv / on (web) Publishing site
- 3 Experimental Design, Overview, and Discussion
- How Ethical Should AI Be? How AI Alignment Shapes the Risk Preferences of LLMs / 2406.01168 / ISBN:https://doi.org/10.48550/arXiv.2406.01168 / Published by ArXiv / on (web) Publishing site
- Introduction
II. Risk Characteristics of LLMs
Figures and tables - Evaluating AI fairness in credit scoring with the BRIO tool / 2406.03292 / ISBN:https://doi.org/10.48550/arXiv.2406.03292 / Published by ArXiv / on (web) Publishing site
- 4 Fairness violation analysis in BRIO
5 Risk assessment in BRIO
6 Risk analysis via BRIO for the German Credit Dataset - MoralBench: Moral Evaluation of LLMs / 2406.04428 / Published by ArXiv / on (web) Publishing site
- Abstract
- Can Prompt Modifiers Control Bias? A Comparative Analysis of Text-to-Image Generative Models / 2406.05602 / Published by ArXiv / on (web) Publishing site
- 1. Introduction
2. Related Work
3. Bias Evaluation
4. Methodology
5. Results
6. Discussion
7. Conclusion - Deception Analysis with Artificial Intelligence: An Interdisciplinary Perspective / 2406.05724 / ISBN:https://doi.org/10.48550/arXiv.2406.05724 / Published by ArXiv / on (web) Publishing site
- 2 Theories and Components of Deception
- The Impact of AI on Academic Research and Publishing / 2406.06009 / Published by ArXiv / on (web) Publishing site
- Ethics of AI for Writing Papers
- An Empirical Design Justice Approach to Identifying Ethical Considerations in the Intersection of Large Language Models and Social Robotics / 2406.06400 / ISBN:https://doi.org/10.48550/arXiv.2406.06400 / Published by ArXiv / on (web) Publishing site
- 4 Findings
5 Discussion - The Ethics of Interaction: Mitigating Security Threats in LLMs / 2401.12273 / ISBN:https://doi.org/10.48550/arXiv.2401.12273 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
3 Potential Misuse and Security Concerns
4 Towards Ethical Mitigation: A Proposed Methodology
5 Preemptive Ethical Measures
6 Ethical Response to LLM Attacks - Global AI Governance in Healthcare: A Cross-Jurisdictional Regulatory Analysis / 2406.08695 / ISBN:https://doi.org/10.48550/arXiv.2406.08695 / Published by ArXiv / on (web) Publishing site
- 4 Global Regulatory Landscape of AI
A Supplemental Tables - Fair by design: A sociotechnical approach to justifying the fairness of AI-enabled systems across the lifecycle / 2406.09029 / ISBN:https://doi.org/10.48550/arXiv.2406.09029 / Published by ArXiv / on (web) Publishing site
- 3 Assuring fairness across the AI lifecycle
4 Assuring AI fairness in healthcare - Some things never change: how far generative AI can really change software engineering practice / 2406.09725 / ISBN:https://doi.org/10.48550/arXiv.2406.09725 / Published by ArXiv / on (web) Publishing site
- 3 Methodology
4 Results - Applications of Generative AI in Healthcare: algorithmic, ethical, legal and societal considerations / 2406.10632 / ISBN:https://doi.org/10.48550/arXiv.2406.10632 / Published by ArXiv / on (web) Publishing site
- II. Selection of application
III. Analysis
Appendix C Algorithmic / technical aspects - Conversational Agents as Catalysts for Critical Thinking: Challenging Design Fixation in Group Design / 2406.11125 / ISBN:https://doi.org/10.48550/arXiv.2406.11125 / Published by ArXiv / on (web) Publishing site
- 2 BEYOND RECOMMENDATIONS:
ENHANCING CRITICAL THINKING WITH
GENERATIVE AI
3 CHALLENGES AND OPPORTUNITIES OF USING CONVERSATIONAL AGENTS IN GROUP DESIGN
4 POTENTIAL SCENARIO AND APPLICATIONS OF CONVERSATIONAL AGENTS IN GROUP DESIGN PROCESS - Current state of LLM Risks and AI Guardrails / 2406.12934 / ISBN:https://doi.org/10.48550/arXiv.2406.12934 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
2 Large Language Model Risks
3 Strategies in Securing Large Language models
5 Open Source Tools
References - Leveraging Large Language Models for Patient Engagement: The Power of Conversational AI in Digital Health / 2406.13659 / ISBN:https://doi.org/10.48550/arXiv.2406.13659 / Published by ArXiv / on (web) Publishing site
- II. RECENT ADVANCEMENTS IN LARGE LANGUAGE
MODELS
IV. DISCUSSION AND F UTURE D IRECTIONS - Documenting Ethical Considerations in Open Source AI Models / 2406.18071 / ISBN:https://doi.org/10.48550/arXiv.2406.18071 / Published by ArXiv / on (web) Publishing site
- 4 RESULTS
- AI Alignment through Reinforcement Learning from Human Feedback? Contradictions and Limitations / 2406.18346 / ISBN:https://doi.org/10.48550/arXiv.2406.18346 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
2 Background
3 Limitations of RLxF
4 The Internal Tensions and Ethical Issues in RLxF
5 Rebooting Safety and Alignment: Integrating AI Ethics and System Safety
6 Conclusion - A Survey on Privacy Attacks Against Digital Twin Systems in AI-Robotics / 2406.18812 / ISBN:https://doi.org/10.48550/arXiv.2406.18812 / Published by ArXiv / on (web) Publishing site
- II. BACKGROUND
III. ATTACKS ON DT-INTEGRATED AI ROBOTS
IV. DT-INTEGRATED ROBOTICS DESIGN CONSIDERATIONS AND DISCUSSION
REFERENCES - Staying vigilant in the Age of AI: From content generation to content authentication / 2407.00922 / ISBN:https://doi.org/10.48550/arXiv.2407.00922 / Published by ArXiv / on (web) Publishing site
- Emphasizing Reasoning Over Detection
- SecGenAI: Enhancing Security of Cloud-based Generative AI Applications within Australian Critical Technologies of National Interest / 2407.01110 / ISBN:https://doi.org/10.48550/arXiv.2407.01110 / Published by ArXiv / on (web) Publishing site
- I.
INTRODUCTION
II. UNDERSTANDING GENAI SECURITY
III. CRITICAL ANALYSIS
IV. SECGENAI FRAMEWORK REQUIREMENTS SPECIFICATIONS - A Blueprint for Auditing Generative AI / 2407.05338 / ISBN:https://doi.org/10.48550/arXiv.2407.05338 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
2 Why audit generative AI systems?
6 Application audits
7 Clarifications and limitations
8 Conclusion - Operationalising AI governance through ethics-based auditing: An industry case study / 2407.06232 / Published by ArXiv / on (web) Publishing site
- 4. An ‘ethics-based’ AI audit
6. Lessons learned from AstraZeneca’s 2021 AI audit - Auditing of AI: Legal, Ethical and Technical Approaches / 2407.06235 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
4 Auditing of AI’s multidisciplinary foundations - FUTURE-AI: International consensus guideline for trustworthy and deployable artificial intelligence in healthcare / 2309.12325 / ISBN:https://doi.org/10.48550/arXiv.2309.12325 / Published by ArXiv / on (web) Publishing site
- REFERENCES:
Table 1 - Rethinking Machine Ethics -- Can LLMs Perform Moral Reasoning through the Lens of Moral Theories? / 2308.15399 / ISBN:https://doi.org/10.48550/arXiv.2308.15399 / Published by ArXiv / on (web) Publishing site
- B Details of Instructions
- Bridging the Global Divide in AI Regulation: A Proposal for a Contextual, Coherent, and Commensurable Framework / 2303.11196 / ISBN:https://doi.org/10.48550/arXiv.2303.11196 / Published by ArXiv / on (web) Publishing site
- II. Global Divide in AI Regulation: Horizontally. Context-Specific
IV. Proposing an Alternative 3C Framework - CogErgLLM: Exploring Large Language Model Systems Design Perspective Using Cognitive Ergonomics / 2407.02885 / ISBN:https://doi.org/10.48550/arXiv.2407.02885 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
4 Design Framework
6 Discussion
References - Generative AI for Health Technology Assessment: Opportunities, Challenges, and Policy Considerations / 2407.11054 / ISBN:https://doi.org/10.48550/arXiv.2407.11054 / Published by ArXiv / on (web) Publishing site
- Applications of generative AI in literature reviews and evidence synthesis
Applications of generative AI to real-world evidence (RWE):
Limitations of generative AI in HTA applications
Glossary
Appendices - Thorns and Algorithms: Navigating Generative AI Challenges Inspired by Giraffes and Acacias / 2407.11360 / ISBN:https://doi.org/10.48550/arXiv.2407.11360 / Published by ArXiv / on (web) Publishing site
- 4 Generative AI and Humans: Risks and Mitigation
6 Discussion
7 Recommendations: Fixing Gen AI’s Value Alignment - Prioritizing High-Consequence Biological Capabilities in Evaluations of Artificial Intelligence Models / 2407.13059 / ISBN:https://doi.org/10.48550/arXiv.2407.13059 / Published by ArXiv / on (web) Publishing site
- Introduction
- Assurance of AI Systems From a Dependability Perspective / 2407.13948 / ISBN:https://doi.org/10.48550/arXiv.2407.13948 / Published by ArXiv / on (web) Publishing site
- 2 Assurance for Systems Extended with AI and ML
3 Assurance of AI Systems for Specific Functions
4 Assurance for General-Purpose AI
5 Assurance and Alignment for AGI - Open Artificial Knowledge / 2407.14371 / ISBN:https://doi.org/10.48550/arXiv.2407.14371 / Published by ArXiv / on (web) Publishing site
- Appendices
- Honest Computing: Achieving demonstrable data lineage and provenance for driving data and process-sensitive policies / 2407.14390 / ISBN:https://doi.org/10.48550/arXiv.2407.14390 / Published by ArXiv / on (web) Publishing site
- 1. Introduction
- RogueGPT: dis-ethical tuning transforms ChatGPT4 into a Rogue AI in 158 Words / 2407.15009 / ISBN:https://doi.org/10.48550/arXiv.2407.15009 / Published by ArXiv / on (web) Publishing site
- I. Introduction
II. Background
III. Methodology
IV. Results
VI. Discussion
VII. Conclusion - Mapping the individual, social, and biospheric impacts of Foundation Models / 2407.17129 / ISBN:https://doi.org/10.48550/arXiv.2407.17129 / Published by ArXiv / on (web) Publishing site
- Abstract
1 Introduction
5 Discussion: Grappling with the Scale and Interconnectedness of Foundation Models
A Appendix - Exploring the Role of Social Support when Integrating Generative AI into Small Business Workflows / 2407.21404 / ISBN:https://doi.org/10.48550/arXiv.2407.21404 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
2 Related Work
4 Findings
6 Discussion and Future Work - Integrating ESG and AI: A Comprehensive Responsible AI Assessment Framework / 2408.00965 / ISBN:https://doi.org/10.48550/arXiv.2408.00965 / Published by ArXiv / on (web) Publishing site
- 3 Methodology
- Surveys Considered Harmful? Reflecting on the Use of Surveys in AI Research, Development, and Governance / 2408.01458 / ISBN:https://doi.org/10.48550/arXiv.2408.01458 / Published by ArXiv / on (web) Publishing site
- B Additional Materials for Pilot Survey
- Improving Large Language Model (LLM) fidelity through context-aware grounding: A systematic approach to reliability and veracity / 2408.04023 / ISBN:https://doi.org/10.48550/arXiv.2408.04023 / Published by ArXiv / on (web) Publishing site
- 1. Introduction
3. Proposed framework
7. Conclusion and Future Directions - Criticizing Ethics According to Artificial Intelligence / 2408.04609 / ISBN:https://doi.org/10.48550/arXiv.2408.04609 / Published by ArXiv / on (web) Publishing site
- 2 Clarifying conceptual ambiguities
- Between Copyright and Computer Science: The Law and Ethics of Generative AI / 2403.14653 / ISBN:https://doi.org/10.48550/arXiv.2403.14653 / Published by ArXiv / on (web) Publishing site
- I. The Why and How Behind LLMs
II. The Difference Between Academic and Commercial Research
III. A Guide for Data in LLM Research
IV. The Path Ahead - The Responsible Foundation Model Development Cheatsheet: A Review of Tools & Resources / 2406.16746 / ISBN:https://doi.org/10.48550/arXiv.2406.16746 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
7 Environmental Impact
8 Model Evaluation
9 Model Release & Monitoring
References - Recent Advances in Generative AI and Large Language Models: Current Status, Challenges, and Perspectives / 2407.14962 / ISBN:https://doi.org/10.48550/arXiv.2407.14962 / Published by ArXiv / on (web) Publishing site
- II. Generative AI
III. Language Modeling
IV. Challenges of Generative AI and LLMs
V. Bridging Research Gaps and Future Directions - VersusDebias: Universal Zero-Shot Debiasing for Text-to-Image Models via SLM-Based Prompt Engineering and Generative Adversary / 2407.19524 / ISBN:https://doi.org/10.48550/arXiv.2407.19524 / Published by ArXiv / on (web) Publishing site
- 3 Method
Appendices - Speculations on Uncertainty and Humane Algorithms / 2408.06736 / ISBN:https://doi.org/10.48550/arXiv.2408.06736 / Published by ArXiv / on (web) Publishing site
- Abstract
1 Introduction
2 The Numbers of the Future
3 Uncertainty Ex Machina
4 Conclusions - Neuro-Symbolic AI for Military Applications / 2408.09224 / ISBN:https://doi.org/10.48550/arXiv.2408.09224 / Published by ArXiv / on (web) Publishing site
- I. Introduction
II. Neuro-Symbolic AI - Don't Kill the Baby: The Case for AI in Arbitration / 2408.11608 / ISBN:https://doi.org/10.48550/arXiv.2408.11608 / Published by ArXiv / on (web) Publishing site
- 1. What is AI
1. Resistance Against AI Does Not Offer Conclusive Reasons for Outright Rejection - CIPHER: Cybersecurity Intelligent Penetration-testing Helper for Ethical Researcher / 2408.11650 / ISBN:https://doi.org/10.48550/arXiv.2408.11650 / Published by ArXiv / on (web) Publishing site
- 2. Background and Related Works
3. Methodology
4. Experiment Results - Promises and challenges of generative artificial intelligence for human learning / 2408.12143 / ISBN:https://doi.org/10.48550/arXiv.2408.12143 / Published by ArXiv / on (web) Publishing site
- 3 Challenges
Tables - Catalog of General Ethical Requirements for AI Certification / 2408.12289 / ISBN:https://doi.org/10.48550/arXiv.2408.12289 / Published by ArXiv / on (web) Publishing site
- 6 Fairness (F)
8 Safety and Robustness (SR)
9 Sustainability (SU)
10 Transparency and Explainability (T)
11 Truthfulness (TR) - Dataset | Mindset = Explainable AI | Interpretable AI / 2408.12420 / ISBN:https://doi.org/10.48550/arXiv.2408.12420 / Published by ArXiv / on (web) Publishing site
- 1. Introduction
2. Literature Review - Is Generative AI the Next Tactical Cyber Weapon For Threat Actors? Unforeseen Implications of AI Generated Cyber Attacks / 2408.12806 / ISBN:https://doi.org/10.48550/arXiv.2408.12806 / Published by ArXiv / on (web) Publishing site
- IV. Attack Methodology
- Has Multimodal Learning Delivered Universal Intelligence in Healthcare? A Comprehensive Survey / 2408.12880 / ISBN:https://doi.org/10.48550/arXiv.2408.12880 / Published by ArXiv / on (web) Publishing site
- 2 Preliminaries
4 Contrastice Foundation Models (CFMs)
5 Multimodal LLMs (MLLMs)
7 Challenges and Future Directions - Aligning XAI with EU Regulations for Smart Biomedical Devices: A Methodology for Compliance Analysis / 2408.15121 / ISBN:https://doi.org/10.48550/arXiv.2408.15121 / Published by ArXiv / on (web) Publishing site
- 3 Methodology
4 Background
5 Explanation Requirements and Legal Explanatory Goals
6 A Categorisation of XAI in Terms of Explanatory Goals
7 Case Studies: Closed-Loop and Semi-Closed-Loop Control
References - Trustworthy and Responsible AI for Human-Centric Autonomous Decision-Making Systems / 2408.15550 / ISBN:https://doi.org/10.48550/arXiv.2408.15550 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
2 Trustworthy and Responsible AI Definition
4 Biases - A Survey for Large Language Models in Biomedicine / 2409.00133 / ISBN:https://doi.org/10.48550/arXiv.2409.00133 / Published by ArXiv / on (web) Publishing site
- Abstract
2 Background
5 Discussion - Digital Homunculi: Reimagining Democracy Research with Generative Agents / 2409.00826 / ISBN:https://doi.org/10.48550/arXiv.2409.00826 / Published by ArXiv / on (web) Publishing site
- 3. How GenAI Could Make a Difference
4. Risks and Caveats
5. Annoyances or Dealbreakers? - AI Governance in Higher Education: Case Studies of Guidance at Big Ten Universities / 2409.02017 / ISBN:https://doi.org/10.48550/arXiv.2409.02017 / Published by ArXiv / on (web) Publishing site
- Methods
Results - Preliminary Insights on Industry Practices for Addressing Fairness Debt / 2409.02432 / ISBN:https://doi.org/10.48550/arXiv.2409.02432 / Published by ArXiv / on (web) Publishing site
- Abstract
4 Findings
6 Conclusions - DetoxBench: Benchmarking Large Language Models for Multitask Fraud & Abuse Detection / 2409.06072 / ISBN:https://doi.org/10.48550/arXiv.2409.06072 / Published by ArXiv / on (web) Publishing site
- 4 LLM Services (Infrastructure)
5 Prompting
6 Results
10 Appendix - Trust and ethical considerations in a multi-modal, explainable AI-driven chatbot tutoring system: The case of collaboratively solving Rubik's Cubeà / 2402.01760 / ISBN:https://doi.org/10.48550/arXiv.2402.01760 / Published by ArXiv / on (web) Publishing site
- D. CausalRating: A Tool To Rate Sentiments Analysis Systems for Bias
- The Responsible Foundation Model Development Cheatsheet: A Review of Tools & Resources / 2406.16746 / ISBN:https://doi.org/10.48550/arXiv.2406.16746 / Published by ArXiv / on (web) Publishing site
- B Cheatsheet Samples
- On the Creativity of Large Language Models / 2304.00008 / ISBN:https://doi.org/10.48550/arXiv.2304.00008 / Published by ArXiv / on (web) Publishing site
- Abstract
2 A Creative Journey from Ada Lovelace to Foundation Models
3 Large Language Models and Boden’s Three Criteria
4 Easy and Hard Problems in Machine Creativity
5 Practical Implications
6 Conclusion - Artificial intelligence to advance Earth observation: : A review of models, recent trends, and pathways forward / 2305.08413 / ISBN:https://doi.org/10.48550/arXiv.2305.08413 / Published by ArXiv / on (web) Publishing site
- Part I Modelling - Machine learning, computer vision
and processing
1
Machine learning and computer vision for Earth observation
4 Explainable AI and causal inference
5 Physics-aware machine learning
References - Why business adoption of quantum and AI technology must be ethical / 2312.10081 / ISBN:https://doi.org/10.48550/arXiv.2312.10081 / Published by ArXiv / on (web) Publishing site
- Argument for acknowledging complexity: the case for
individual ethos, regulation is not enough
- Data-Centric Foundation Models in Computational Healthcare: A Survey / 2401.02458 / ISBN:https://doi.org/10.48550/arXiv.2401.02458 / Published by ArXiv / on (web) Publishing site
- 2 Foundation Models
4 Multi-Modal Data Fusion
7 Data Privacy
8 Performance Evaluation
9 Challenges and Opportunities
A Healthcare Data Modalities - Integrating Generative AI in Hackathons: Opportunities, Challenges, and Educational Implications / 2401.17434 / ISBN:https://doi.org/10.48550/arXiv.2401.17434 / Published by ArXiv / on (web) Publishing site
- 4. Discussion
- Large language models as linguistic simulators and cognitive models in human research / 2402.04470 / ISBN:https://doi.org/10.48550/arXiv.2402.04470 / Published by ArXiv / on (web) Publishing site
- Language models as human participants
Six fallacies that misinterpret language models
Using language models to simulate roles and model cognitive processes - Navigating LLM Ethics: Advancements, Challenges, and Future Directions / 2406.18841 / ISBN:https://doi.org/10.48550/arXiv.2406.18841 / Published by ArXiv / on (web) Publishing site
- IV. Findings and Resultant Themes
V. Discussion - How Mature is Requirements Engineering for AI-based Systems? A Systematic Mapping Study on Practices, Challenges, and Future Research Directions / 2409.07192 / ISBN:https://doi.org/10.48550/arXiv.2409.07192 / Published by ArXiv / on (web) Publishing site
- 4 Results
5 Open Challenges and Future Research Directions (RQ5)
References - Improving governance outcomes through AI documentation: Bridging theory and practice / 2409.08960 / ISBN:https://doi.org/10.48550/arXiv.2409.08960 / Published by ArXiv / on (web) Publishing site
- 4 Results
- ValueCompass: A Framework of Fundamental Values for Human-AI Alignment / 2409.09586 / ISBN:https://doi.org/10.48550/arXiv.2409.09586 / Published by ArXiv / on (web) Publishing site
- 2 Related Work
4 Operationalizing ValueCompass: Methods to Measure Value Alignment of Humans and AI
5 Findings with ValueCompass: The Status Quo of Human-AI Value Alignment
6 Discussion - Reporting Non-Consensual Intimate Media: An Audit Study of Deepfakes / 2409.12138 / ISBN:https://doi.org/10.48550/arXiv.2409.12138 / Published by ArXiv / on (web) Publishing site
- 2 Related Research
- Generative AI Carries Non-Democratic Biases and Stereotypes: Representation of Women, Black Individuals, Age Groups, and People with Disability in AI-Generated Images across Occupations / 2409.13869 / ISBN:https://doi.org/10.48550/arXiv.2409.13869 / Published by ArXiv / on (web) Publishing site
- Abstract
Mutual Impacts: Technology and Democracy
Data and Results - GenAI Advertising: Risks of Personalizing Ads with LLMs / 2409.15436 / ISBN:https://doi.org/10.48550/arXiv.2409.15436 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
2 Background and Related Work - XTRUST: On the Multilingual Trustworthiness of Large Language Models / 2409.15762 / ISBN:https://doi.org/10.48550/arXiv.2409.15762 / Published by ArXiv / on (web) Publishing site
- 4 Experiments
- Artificial Human Intelligence: The role of Humans in the Development of Next Generation AI / 2409.16001 / ISBN:https://doi.org/10.48550/arXiv.2409.16001 / Published by ArXiv / on (web) Publishing site
- 4 Brain-inspired Information processing
5 Challenges and Perspectives in Human-Level AI Development
References - Ethical and Scalable Automation: A Governance and Compliance Framework for Business Applications / 2409.16872 / ISBN:https://doi.org/10.48550/arXiv.2409.16872 / Published by ArXiv / on (web) Publishing site
- 2. Literature Review
3. Methodology
4. Framework Development
5. Analysis and Discussion - Decoding Large-Language Models: A Systematic Overview of Socio-Technical Impacts, Constraints, and Emerging Questions / 2409.16974 / ISBN:https://doi.org/10.48550/arXiv.2409.16974 / Published by ArXiv / on (web) Publishing site
- 6 Methodologies & Capabilities (RQ2)
7 Limitations & Considerations (RQ3)
8 Discussion
References - Safety challenges of AI in medicine / 2409.18968 / ISBN:https://doi.org/10.48550/arXiv.2409.18968 / Published by ArXiv / on (web) Publishing site
- 2 Inherent problems of AI related to medicine
3 Risks of using AI in medicine - Responsible AI in Open Ecosystems: Reconciling Innovation with Risk Assessment and Disclosure / 2409.19104 / ISBN:https://doi.org/10.48550/arXiv.2409.19104 / Published by ArXiv / on (web) Publishing site
- 2 Related Work
- AI-Press: A Multi-Agent News Generating and Feedback Simulation System Powered by Large Language Models / 2410.07561 / ISBN:https://doi.org/10.48550/arXiv.2410.07561 / Published by ArXiv / on (web) Publishing site
- 2 Related Works
3 AI Press System
A User Interface
F User Profile Pool Generating
G Evaluation Experiment - DailyDilemmas: Revealing Value Preferences of LLMs with Quandaries of Daily Life / 2410.02683 / ISBN:https://doi.org/10.48550/arXiv.2410.02683 / Published by ArXiv / on (web) Publishing site
- Appendices
- From human-centered to social-centered artificial intelligence: Assessing ChatGPT's impact through disruptive events / 2306.00227 / ISBN:https://doi.org/10.48550/arXiv.2306.00227 / Published by ArXiv / on (web) Publishing site
- Introduction
The multiple levels of AI impact
The emerging social impacts of ChatGPT
Discussion - The Design Space of in-IDE Human-AI Experience / 2410.08676 / ISBN:https://doi.org/10.48550/arXiv.2410.08676 / Published by ArXiv / on (web) Publishing site
- II. Related Work
IV. Results
V. Discussion
VII. Conclusion - Learning Human-like Representations to Enable Learning Human Values / 2312.14106 / ISBN:https://doi.org/10.48550/arXiv.2312.14106 / Published by ArXiv / on (web) Publishing site
- A. Appendix
- When AI Eats Itself: On the Caveats of Data Pollution in the Era of Generative AI / 2405.09597 / ISBN:https://doi.org/10.48550/arXiv.2405.09597 / Published by ArXiv / on (web) Publishing site
- References
- Study on the Helpfulness of Explainable Artificial Intelligence / 2410.11896 / ISBN:https://doi.org/10.48550/arXiv.2410.11896 / Published by ArXiv / on (web) Publishing site
- 2 Measuring Explainability
3 An objective Methodology for evaluating XAI - Navigating the Cultural Kaleidoscope: A Hitchhiker's Guide to Sensitivity in Large Language Models / 2410.12880 / ISBN:https://doi.org/10.48550/arXiv.2410.12880 / Published by ArXiv / on (web) Publishing site
- Abstract
1 Introduction
2 Previous studies
4 Cultural safety dataset
5 Experimental setup
7 Cultural safeguarding
9 Conclusion - How Do AI Companies Fine-Tune Policy? Examining Regulatory Capture in AI Governance / 2410.13042 / ISBN:https://doi.org/10.48550/arXiv.2410.13042 / Published by ArXiv / on (web) Publishing site
- Executive Summary
2 Defining “Regulatory Capture” - Data Defenses Against Large Language Models / 2410.13138 / ISBN:https://doi.org/10.48550/arXiv.2410.13138 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
2 Ethics of Resisting LLM Inference
3 Threat Model
4 LLM Adversarial Attacks as LLM Inference Data Defenses - Do LLMs Have Political Correctness? Analyzing Ethical Biases and Jailbreak Vulnerabilities in AI Systems / 2410.13334 / ISBN:https://doi.org/10.48550/arXiv.2410.13334 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
2 Background and Related Works
3 Methodology PCJAILBREAK
5 Conclusion
Refefences - A Simulation System Towards Solving Societal-Scale Manipulation / 2410.13915 / ISBN:https://doi.org/10.48550/arXiv.2410.13915 / Published by ArXiv / on (web) Publishing site
- 4 Analysis
- Confrontation or Acceptance: Understanding Novice Visual Artists' Perception towards AI-assisted Art Creation / 2410.14925 / ISBN:https://doi.org/10.48550/arXiv.2410.14925 / Published by ArXiv / on (web) Publishing site
- 2 Background and Related Work
5 RQ1: Evolution of the Opinions Towards AI Tools - Jailbreaking and Mitigation of Vulnerabilities in Large Language Models / 2410.15236 / ISBN:https://doi.org/10.48550/arXiv.2410.15236 / Published by ArXiv / on (web) Publishing site
- I. Introduction
II. Background and Concepts
III. Jailbreak Attack Methods and Techniques
IV. Defense Mechanisms Against Jailbreak Attacks
VI. Research Gaps and Future Directions
VII. Conclusion
References - Redefining Finance: The Influence of Artificial Intelligence (AI) and Machine Learning (ML) / 2410.15951 / ISBN:https://doi.org/10.48550/arXiv.2410.15951 / Published by ArXiv / on (web) Publishing site
- What Is AI & ML
- Vernacularizing Taxonomies of Harm is Essential for Operationalizing Holistic AI Safety / 2410.16562 / ISBN:https://doi.org/10.48550/arXiv.2410.16562 / Published by ArXiv / on (web) Publishing site
- Taxonomies of Harm Must be Vernacularized to be Operationalized
- Trustworthy XAI and Application / 2410.17139 / ISBN:https://doi.org/10.48550/arXiv.2410.17139 / Published by ArXiv / on (web) Publishing site
- 1 Introduction
3 Applications of Trustworthy XAI - Towards Automated Penetration Testing: Introducing LLM Benchmark, Analysis, and Improvements / 2410.17141 / ISBN:https://doi.org/10.48550/arXiv.2410.17141 / Published by ArXiv / on (web) Publishing site
- 4 Evaluation
5 Discussion
Supplementary Materials - Demystifying Large Language Models for Medicine: A Primer / 2410.18856 / ISBN:https://doi.org/10.48550/arXiv.2410.18856 / Published by ArXiv / on (web) Publishing site
- Introduction
Task Formulation
Large Language Model Selection
Prompt engineering
Deployment considerations
Glossary - The Cat and Mouse Game: The Ongoing Arms Race Between Diffusion Models and Detection Methods / 2410.18866 / ISBN:https://doi.org/10.48550/arXiv.2410.18866 / Published by ArXiv / on (web) Publishing site
- II. Fundamentals of Diffusion Models and Detection Challenges
VI. Evaluation Metrics
VII. Applications and Implications - TRIAGE: Ethical Benchmarking of AI Models Through Mass Casualty Simulations / 2410.18991 / ISBN:https://doi.org/10.48550/arXiv.2410.18991 / Published by ArXiv / on (web) Publishing site
- Appendices
- My Replika Cheated on Me and She Liked It: A Taxonomy of Algorithmic Harms in Human-AI Relationships / 2410.20130 / ISBN:https://doi.org/10.48550/arXiv.2410.20130 / Published by ArXiv / on (web) Publishing site
- 3 Methodology
5 Discussion - Democratizing Reward Design for Personal and Representative Value-Alignment / 2410.22203 / ISBN:https://doi.org/10.48550/arXiv.2410.22203 / Published by ArXiv / on (web) Publishing site
- 3 Interactive-Reflective Dialogue Alignment (IRDA) System
Appendices - Moral Agency in Silico: Exploring Free Will in Large Language Models / 2410.23310 / ISBN:https://doi.org/10.48550/arXiv.2410.23310 / Published by ArXiv / on (web) Publishing site
- Introduction
Defining Key Concepts
Theoretical Framework
Discussion - Web Scraping for Research: Legal, Ethical, Institutional, and Scientific Considerations / 2410.23432 / ISBN:https://doi.org/10.48550/arXiv.2410.23432 / Published by ArXiv / on (web) Publishing site
- 3 Research Considerations
- Using Large Language Models for a standard assessment mapping for sustainable communities / 2411.00208 / ISBN:https://doi.org/10.48550/arXiv.2411.00208 / Published by ArXiv / on (web) Publishing site
- 3 Methodology
4 CaseStudies and Results
Acknowledgement - Where Assessment Validation and Responsible AI Meet / 2411.02577 / ISBN:https://doi.org/10.48550/arXiv.2411.02577 / Published by ArXiv / on (web) Publishing site
- Classical Assessment Validation Theory and Responsible AI
The Evolution of Responsible AI for Assessment
Integrating Classical Validation Theory and Responsible AI - Examining Human-AI Collaboration for Co-Writing Constructive Comments Online / 2411.03295 / ISBN:https://doi.org/10.48550/arXiv.2411.03295 / Published by ArXiv / on (web) Publishing site
- 3 Methods
4 Findings
5 Discussion
A Appendix - I Always Felt that Something Was Wrong.: Understanding Compliance Risks and Mitigation Strategies when Professionals Use Large Language Models / 2411.04576 / ISBN:https://doi.org/10.48550/arXiv.2411.04576 / Published by ArXiv / on (web) Publishing site
- 2 Background and Related Work
3 Method: Semi-structured Interviews
Appendices - Navigating the Cultural Kaleidoscope: A Hitchhiker's Guide to Sensitivity in Large Language Models / 2410.12880 / ISBN:https://doi.org/10.48550/arXiv.2410.12880 / Published by ArXiv / on (web) Publishing site
- Appendices