To Build or Not to Build? Factors that Lead to Non-Development or Abandonment of AI Systems
A research paper investigates factors that lead organizations to abandon AI systems during development or post-deployment, finding that ethical concerns represent only one of six drivers. The study reveals that practical constraints—including resource limitations, organizational dynamics, and regulatory pressures—often outweigh ethical considerations in non-development decisions, suggesting responsible AI research should broaden its focus beyond ethics-centric approaches.
This academic research addresses a significant blind spot in responsible AI discourse: why organizations choose not to build or deploy AI systems. Rather than assuming ethical reasoning dominates abandonment decisions, the authors conducted a comprehensive review combining literature analysis, incident databases, and practitioner surveys to identify real-world drivers. The taxonomy identifies six distinct categories: ethical concerns, stakeholder feedback, development lifecycle challenges, organizational dynamics, resource constraints, and legal/regulatory concerns.
The research fills an important gap because most responsible AI literature examines deployed systems' impacts rather than pre-deployment decision-making. By investigating the earlier development stages, this work identifies intervention points that could prevent problematic systems from reaching deployment without requiring external regulation. The finding that non-ethics-related factors often dominate abandonment decisions has significant implications for how researchers and practitioners approach AI governance.
For the AI development ecosystem, this research suggests current responsible AI frameworks may overemphasize philosophical and ethical arguments while underestimating practical levers like cost-benefit analysis, team constraints, and market feedback. Organizations making build-or-abandon decisions respond to diverse pressures; practitioners seeking to discourage harmful AI development might achieve greater success by engaging with economic, operational, and organizational factors rather than relying solely on ethical arguments.
Looking forward, this work creates space for responsible AI research to develop more nuanced strategies that account for how organizations actually make decisions. Future research should examine whether these abandonment levers can be intentionally strengthened—through policy, market mechanisms, or institutional design—to prevent harmful AI development before deployment occurs.
- →Ethical concerns represent only one of six factors driving AI system abandonment; practical constraints often prove more influential.
- →Pre-deployment decisions shape which AI systems reach users, representing underexplored intervention points for responsible AI governance.
- →Organizations abandon AI development for diverse reasons including resource constraints, organizational dynamics, regulatory concerns, and stakeholder feedback.
- →Current responsible AI research emphasizes ethics-centric approaches but may neglect economic and operational levers that actually influence abandonment decisions.
- →Responsible AI strategies could improve effectiveness by addressing practical development barriers rather than relying primarily on ethical arguments.