By Andrew Miller
Three years ago, about one in five small businesses used AI. Today that number is closer to three out of four, according to the House Small Business Committee.
Committee Chair Roger Williams pointed out that 82% of the small businesses using AI grew their workforce over the past year. The upside is real. Although that’s not the story usually told about automation and jobs. AI tools are being implemented alongside existing workers, not simply replacing them.
Even though AI has seen widespread adoption among small businesses, many aren’t equipped to use it efficiently and safely. That was the topic of discussion during the Committee’s hearing last week.
One such concern is what happens when an AI gets something wrong in front of a customer. That question dates back to 2024, when a Vancouver man named Jake Moffatt was booking a last-minute flight to attend his grandmother’s funeral.
After asking Air Canada’s website chatbot about bereavement fares, it told him to book the regular fare and apply for the discount within 90 days after flying. When he did just that, Air Canada denied the claim. Its actual policy, not the one falsely stated by AI, required approval before travel, not after.
The case was taken to British Columbia’s Civil Resolution Tribunal. Air Canada argued that the chatbot was a separate entity, and that the correct policy was posted elsewhere on the airline’s own site. The tribunal rejected both arguments and ordered the airline to pay Moffatt the fare difference.
The ruling is regularly used in similar disputes. Essentially, whatever your AI tells a customer, your business said it. Small businesses building AI into their websites and phone lines carry that same exposure. If a chatbot invents a return policy or quotes the wrong price, the business is responsible for making it right. It doesn’t fall on the software vendor that built the bot.
That divide isn’t about whether a business is responsible for what its own chatbot tells a customer. Courts already settled that. It’s about how much guidance small businesses get from Washington while they figure out how to use AI safely.
Republicans argued the priority should be removing regulatory friction so AI adoption keeps climbing. Democrats pushed for more SBA-led literacy resources, pointing to hallucinations where AI makes up information as a risk small businesses are underprepared for.
Nothing was decided at the hearing itself. The three bills addressing AI training and literacy passed the House months ago and are still sitting in the Senate.
Until that changes, the Moffatt ruling is the clearest answer a small business has: whatever your AI tells a customer, your business said it, no matter what Washington decides to do next.
P.S. AI can do a lot of things well. Finding you the right financing isn’t one of them yet. That’s still a Signet conversation. Contact us now to explore your options.