In this session, TenForce Head of Product,  Uros Milosevic, explains how AI can support EHS teams. He shares practical use cases, responsible implementation practices, and the choices organizations need to make before introducing AI into safety-critical processes.

You’ll also see how TenForce uses AI to support Incident Management and Permit to Work.

What you’ll learn

Watch the session to understand:

• How AI assistants differ from custom machine-learning solutions
• Which approach is better suited to different EHS tasks
• Why human review remains essential
• How external AI tools can connect to approved EHS data
• What to consider around access, privacy and model choice

Practical AI for the work EHS teams do every day

During the session, you will see how AI can support:

• Incident reporting by turning spoken or written descriptions into clearer reports, even across languages.
• Incident investigation by suggesting possible follow-up actions.
• Hazard identification by bringing overlooked hazards into view.
• Control measures by suggesting relevant controls based on the hazards.
• Corrective and preventive actions by giving teams a stronger starting point when developing action plans after incidents, audits and inspections.
• Reporting and analysis by summarizing information, identifying recurring patterns and explaining findings in clear language.

See AI support inside real EHS workflows

The session also includes demonstrations of two TenForce AI Assistants.

Strengthen incident action plans

See how the AI Assistant for Incident Management reviews an incident description, investigation report, root causes and existing actions to recommend corrective actions.

Bring overlooked hazards into view

See how the AI Assistant for Permit to Work reviews a job description and risk assessment to suggest relevant hazards already held in your master data and identify potential hazards that may not yet be included.

Connect EHS data to the AI tools your business uses

The session also introduces Model Context Protocol, or MCP. Watch the demonstration to see how a connected assistant can analyze incident records, identify recurring themes and recommend a prioritized action plan.

About the speaker

Uros is responsible for the product vision across the TenForce EHSQ platform, modules and mobile application. His background is in artificial intelligence and intelligent information systems. He has coordinated major EU-funded AI research projects and advised European institutions on using machine learning to improve data and processes.