Check jobsite photos for hazards with an AI safety coach
Turner Construction
What they did with AI
Turner's SafeT Coach answers plain-language safety questions using Turner's own EH&S standards and reviews uploaded jobsite photos to flag likely hazards, rank the risk and suggest controls. One superintendent used it to work out whether a vertical shaft counted as a permit-required confined space and got a decision flow chart, a daily permit checklist and policy citations within minutes. The tool has logged 25,000+ interactions, and Turner made it free for the whole industry in May 2026.
Results above are as reported by the vendor or company in the source and haven't been independently audited by this site. Listing a tool is not an endorsement, and nobody paid for this entry.
Related use cases
Generate site-specific toolbox talks with an internal AI tool
Skanska USA Building
Skanska USA built an internal "Safety Sidekick" on generative AI trained on its own safety data and jobsite conditions. Field staff use it on mobile or desktop to plan toolbox talks and morning huddles, ask safety questions, and ask about photos of the day's site conditions. Early users report less time spent preparing safety briefings and better jobsite engagement. Skanska's advice to smaller contractors: start with data you already have, or use an off-the-shelf tool rather than building your own.
Build a custom GPT that writes code-compliant roof repair estimates from a voice prompt
Leakbusters Roofing
Owner Richie Kleti's team built a custom GPT, tied to historical job data and manufacturer specs, that turns a short voice prompt into an itemized, code-compliant Florida repair estimate. New hires can produce accurate estimates without waiting for senior staff to price every job. The team also uses an AI sales coach in role-play "sales cage match" sessions, where reps are mic'd up and get feedback on tone, objection handling and what to do better.
Answer every call 24/7 and score every CSR call with AI
Rescue Air & Plumbing
Rescue Air & Plumbing in Dallas-Fort Worth estimated that each missed call cost it about $250. It put an AI CSR on all inbound calls (pulling caller info, checking tech schedules and booking into the CRM) and uses an AI coaching tool that scores every call so CSRs get feedback on each one instead of a sample. The case study credits the coaching program with two agents being promoted.