AI Safety and Evaluation

Finally, many projects involve some combination of audits and benchmarks.

Key References

  1. 2025 · PaperAn Audit and Analysis of LLM-Assisted Health Misinformation Jailbreaks Against LLMsAIES
  2. 2026 · Other paperIDP-Bench: Benchmarking ability of LLMs to protect personal information in interdependent privacy contextsarXiv preprint

More work in this topic

A not-totally-exhaustive (because there’s a lot of overlap between these themes) list of “outputs” that meet this theme.

2026May 2026PanelPanelist, Responsible AI TrackCanadian AI/CRV 2026
2026May 2026PaperOverreliance in Writing Tasks: Exploring Similarity-Based Measures of AI Influence on Writing and Proposing a Reflective Writing Interface InterventionACM FAccT
2026Apr 2026BlogThe AI "Evaluation Crisis" Is an Opportunity to Get Data Flow RightData Leverage
2025Nov 2025Other paperBenchmarks 101NeurIPS 2025 Education Program
2025Oct 2025Other paperOpen WebUI: An Open, Extensible, and Usable Interface for AI InteractionarXiv preprint
2025Aug 2025TalkData Leverage and AI SafetySingapore AI Safety Hub
2025Jul 2025MediaAI as 'teammate'? Not so fast, say experts warning it could be 'dangerous'Canadian HR Reporter
2025May 2025BlogHow do we know our AI output is good? Double checks, bar charts, vibes, and training data.Data Leverage
2025May 2025Media‘Trial and error’: Over half of AI redundancies a mistake, say leadersCanadian HR Reporter
2025Apr 2025WorkshopStep-By-Step Reasoning with Meta Cognitive Prompts to Reduce Contextual HallucinationHEAL Workshop @ CHI 2025
2024Aug 2024Blog"Many Models" and "Track Changes" for AI: Some Thoughts on LLM InterfacesData Leverage
2024May 2024WorkshopDesigning an open-source LLM interface and social platforms for collectively driven LLM evaluation and auditingHEAL Workshop @ CHI 2024
2024May 2024WorkshopThe Need for Flexible Interfaces for Text-to-Image Auditing: A Case Study of DALL·E 2 and DALL·E 3GenAICHI @ CHI 2024
2024Jan 2024PanelResponsive LLM DevelopmentUniversity of Toronto Data Sciences Institute: Fairness - ChatGPT Workshop
2023Jan 2023TalkAlgorithmic ContestabilityCenter for Advancing Safety of Machine Intelligence, Toward a Safety Science of AI
2015Nov 2015WorkshopDeep learning of tissue fate features in acute ischemic strokeBIBM 2015 Workshop on Biomedical Visual Search and Deep Learning
2015Nov 2015WorkshopDetection of hyperperfusion on arterial spin labeling using deep learningBIBM 2015 Workshop on Biomedical Visual Search and Deep Learning