Safety Index publishes plain-language workplace safety guides. Each one is built from primary sources: the regulation, the standard or the regulator's own guidance, linked in the sources list so you can read the original.
How we write
Guides are researched and written by Claude, an AI model made by Anthropic. During research it opens the regulator and standards pages it cites, and the build refuses a guide that cites a page it did not open. We do not invent authors, case studies, statistics or experience: worked examples are labelled as examples. Featured images are generated by Google's Gemini image model, are illustrative only, and carry the IPTC label for AI-generated media.
Google's guidance is that content is judged on whether it helps people, not on how it was produced. We still tell you plainly how this is made, because safety is a topic where accuracy matters. Guides are not reviewed line by line by a named safety professional.
What to do with a guide
Use it to understand a rule and find the source. Before you rely on it for your own site, read the linked regulation and ask your regulator or a competent safety professional. If you find a mistake, tell us on the contact page and we will correct it and update the date.
Languages
Each guide is also written for French and German readers. These are rewrites for that market, researched with local search terms and local regulators where they exist, not word-for-word translations.
Safety Index is an independent workplace safety blog: plain-language guides to OSHA, ANSI, NFPA and NIOSH requirements, each linked to the primary sources it was written from.