The Many Ways AI is Improving Workplace Safety
Julia Penfield, Ph.D.
August 18, 2025
Simultaneously, artificial intelligence (AI) has matured to the point where it can now support—and in many ways, transform—how safety teams identify, assess, and mitigate risk. With increasing workforce shortages in safety professions, growing regulatory and auditory pressures limiting the available bandwidth of EHS professionals, and rising expectations around corporate responsibility, the integration of AI is not only timely but urgent. EHS professionals can’t afford to ignore AI, because it offers a generational opportunity to shift from a reactive to proactive management approach.
Here are five examples of how AI is being used to improve workplace safety:
- Automated risk identification for Potential Serious Injury and Fatality (PSIF) insights
AI can scan incident reports to flag serious risks that are hiding in the details of less serious incidents, such as near misses, that often go undetected when human oversight falls short. - AI-driven Job Safety Assessments (JSAs)
AI can provide instant, context-aware guidance to improve job descriptions in JSAs and provide guidance on hazards and controls, acting as a virtual safety advisor. - Musculoskeletal disorder (MSD) risk detection
Through video analysis, AI can detect risks of repetitive strain or other MSDs that may otherwise go unnoticed and help identify root causes and effective controls. - Root cause analysis after incidents
AI models can analyze complex incident data to uncover systemic or hidden root causes, improving post-incident learning and prevention. - Enterprise-wide risk insights
AI enables aggregation and analysis of safety data across multiple worksites, breaking down silos and uncovering systemic trends.
- Utilize existing safety data
Identify where risk-related data is collected, how it's stored, and how accessible it is for AI use. - Pilot AI-driven use cases
Start small with targeted applications to demonstrate proof of concept and ROI. - Engage cross-functional teams
Collaborate with IT, operations, and legal to address implementation hurdles, compliance, privacy, and governance requirements. - Invest in AI literacy for safety teams
Provide basic training on how AI works and how it can assist (not replace) safety professionals by helping to increase their bandwidth. - Promote a safety-first AI ethos
Frame AI adoption as the support needed for the EHS professional’s moral and professional mission to get people home safely every day. It’s a way to save lives, not just streamline workflows.
Related
Sponsored Content
About the Author
Julia Penfield, Ph.D.
VelocityEHS
Dr. Penfield has decades of experience in Machine Learning (ML) and Artificial Intelligence (AI) and leads these initiatives across VelocityEHS. She is an expert in predictive modeling, failure prediction, anomaly detection, long-term and short-term forecasting with proper uncertainty estimations, root cause diagnostics, and mathematical optimization, driving the development of advanced technologies that strengthen risk management and operational performance. She also brings experience with Natural Language Processing (NLP) and Computer Vision (CV) and holds seven patents for her innovations. Prior to joining VelocityEHS, she led the machine learning and data sciences team at BC Hydro, one of Canada’s largest energy providers, where she guided AI-driven predictive maintenance and forecasting initiatives. She holds a PhD in the application of machine learning in electrical engineering from the University of British Columbia.