why this matters
general-purpose AI fails WHS practitioners in ways a non-expert never catches: hallucinated section numbers, NSW regulations quoted in a WA context, generic advice where an ICAM investigation was needed. in a safety-critical domain, a confident wrong answer erodes trust and can contribute to poor decisions.
what we're doing
- answers from more than 30,000 passages across 338 documents
- ten jurisdictions, chosen before the question is asked
- every section it cites is checked against its sources and marked found or not found
- where no verified penalty figure is held, the question is declined before the model runs
- fourteen guided tasks, from reviewing a SWMS to preparing a board paper
- questions, documents and chats are never used to train a model
where it's at
SafetyLM is opening to practitioners in small groups, through the waitlist at safetylm.ai.