Why Brokers Try ChatGPT and Walk Away
Talk to brokers who have tried ChatGPT and a pattern appears: they use it for a week, get burned, and quit. The failures are specific — and instructive.
The specific failures
A general tool does not know SPF forms. It can confuse sewer backup with overland flood — a distinction that matters at claim time. And it often writes in American English for Canadian clients, citing coverage concepts from a different regulatory world.
Even good output is work
When the output is usable, the process still costs time: write the prompt, supply the context, verify the answer, and translate it to Canadian — every session. The tool saves writing time and spends it back on babysitting.
Teach it once, or build it in
None of this means AI cannot help brokers. It means a general-purpose tool has to be taught insurance from scratch every time, by someone who already knows the answer. A purpose-built tool flips that: context built in, prompts written, output ready to review and send.
The takeaway: General AI tools do not know Canadian insurance, so brokers spend saved time babysitting output; a purpose-built tool loads the context once so the writing starts finished.