Session
Hasso Plattner Institut
Claire Sophie Patzig, Hasso Plattner Institut, Technical Community, Western European and Others Group
Claire Sophie Patzig, Hasso Plattner Institut, Technical Community, Western European and Others Group
Organization's Website
Speakers
Claire Sophie Patzig, Hasso Plattner Institut, Technical Community, Western European and Others Group
Onsite Moderator
Claire Sophie Patzig
Rapporteur
Claire Sophie Patzig
SDGs
9. Industry, Innovation and Infrastructure
10. Reduced Inequalities
17. Partnerships for the Goals
Targets: This proposal aligns with SDG 9 (Industry, Innovation, and Infrastructure), SDG 10 (Reduced Inequalities), and SDG 17 (Partnerships for the Goals) by addressing fundamental challenges in AI development that impact sustainable technological progress, fairness, and collaboration. AI governance today often focuses on regulating applications rather than addressing the structural barriers that hinder responsible innovation. This talk highlights the need for better data collection mechanisms, improved labeling processes, and ethically sourced datasets—all critical for building a sustainable and equitable AI ecosystem. Strengthening these foundations ensures that AI contributes to inclusive industrial and technological development rather than reinforcing existing inefficiencies. Many AI systems suffer from poor-quality or biased data, leading to unfair and unreliable outcomes. By reducing reliance on low-quality labeling work and promoting diverse, representative datasets, this proposal supports the creation of more inclusive AI systems that work across different populations and contexts. Ensuring equal access to high-quality AI development resources helps mitigate bias and disparities in AI-driven decision-making, particularly in critical sectors like healthcare. The challenges outlined in this proposal cannot be solved in isolation. AI development requires stronger collaboration between developers, policymakers, and ethical data providers. By bridging the gap between governance and real-world AI development, this talk emphasizes the importance of multi-stakeholder partnerships in shaping responsible and effective AI systems. Enabling cooperation between technical and regulatory communities ensures AI can deliver meaningful and equitable progress across industries.
10. Reduced Inequalities
17. Partnerships for the Goals
Targets: This proposal aligns with SDG 9 (Industry, Innovation, and Infrastructure), SDG 10 (Reduced Inequalities), and SDG 17 (Partnerships for the Goals) by addressing fundamental challenges in AI development that impact sustainable technological progress, fairness, and collaboration. AI governance today often focuses on regulating applications rather than addressing the structural barriers that hinder responsible innovation. This talk highlights the need for better data collection mechanisms, improved labeling processes, and ethically sourced datasets—all critical for building a sustainable and equitable AI ecosystem. Strengthening these foundations ensures that AI contributes to inclusive industrial and technological development rather than reinforcing existing inefficiencies. Many AI systems suffer from poor-quality or biased data, leading to unfair and unreliable outcomes. By reducing reliance on low-quality labeling work and promoting diverse, representative datasets, this proposal supports the creation of more inclusive AI systems that work across different populations and contexts. Ensuring equal access to high-quality AI development resources helps mitigate bias and disparities in AI-driven decision-making, particularly in critical sectors like healthcare. The challenges outlined in this proposal cannot be solved in isolation. AI development requires stronger collaboration between developers, policymakers, and ethical data providers. By bridging the gap between governance and real-world AI development, this talk emphasizes the importance of multi-stakeholder partnerships in shaping responsible and effective AI systems. Enabling cooperation between technical and regulatory communities ensures AI can deliver meaningful and equitable progress across industries.
Format
Lightning Talk with focus on challenges in the development process of a project in the field of Digital Health.
Duration (minutes)
20
Description
Many AI governance discussions overlook the realities of AI development, failing to address the most pressing challenges from a developer’s perspective. AI systems do not fail because of faulty code but due to deeper structural issues in the development process. Poor data quality, if usable data exists at all, and an ambiguous legal framework create significant roadblocks. Governance efforts currently focus too much on regulating applications rather than resolving the systemic flaws that undermine responsible AI development.
Drawing from a digital health project involving Electronic Health Records (EHRs), this session highlights governance challenges that extend beyond technical inconveniences and shape whether AI systems are safe, effective, and inclusive. Unclear data governance structures complicate access and usability. Systemic flaws such as redundant documentation practices and coding inconsistencies degrade data quality. Limited interoperability further restricts seamless data integration and large-scale AI applications.
AI governance must go beyond outcome regulation and actively foster the conditions for ethical and effective AI development. Addressing these foundational issues requires collaboration between policymakers, developers, and stakeholders to build AI systems that are not only innovative but also just, inclusive, and trustworthy.
Not applicable for a Lighning Talk.
Not applicable for a Lighning Talk.