Canadian workers are bringing AI into their jobs faster than many employers are explaining how to use it safely. A survey of 1,042 working Canadians found that 48% had never been shown safe-use practices by their employer, while 39% said they used free or personal AI accounts for work.
Watch Desk analysis
What happened
The Angus Reid Forum survey was commissioned by ISA Cybersecurity. Digital Journal reports that 24% of respondents use tools such as ChatGPT, Claude, Copilot or Gemini daily, and 47% use them at least a few times a week. Nearly three-quarters said their AI use had increased moderately or significantly over the past year; two in five plan to rely on it more in 2027.
The gap is not just about employees picking the wrong chatbot. Among daily AI users, one in four said they had put internal documents, client data or code into tools not provided by their employer. And 44% of daily users said they had let an AI tool read email or calendars, open files, send messages or make a purchase.
Why it matters
An AI assistant can only handle information and take actions available to it. If staff do not know which tools are approved, what information can be shared or how much access to grant, an everyday shortcut can expose company data or give software more reach than a task requires.
The article says half of respondents reported a clear workplace AI policy. Another 18% said a policy existed but was unclear, while 17% said there was none. The figures come from a survey commissioned by a cybersecurity company, not a census of Canadian workplaces, but they offer a concrete picture of the training and policy gap.
Our read
“Use AI responsibly” is not a policy if nobody can tell what it means on Tuesday afternoon. Employers need to name approved tools, spell out what staff may share, and review the permissions granted to assistants. They should also give people a quick way to check suspicious requests and report mistakes. Training is less glamorous than another AI launch, but rather more useful when someone is about to paste a client file into the wrong window.
What to watch
- Whether employers turn general AI rules into specific guidance on tools, data and permissions.
- Whether access reviews become a routine part of rolling out AI assistants.
- Whether more training reaches workers as daily workplace AI use grows.
Discussion spark: Should employers be responsible for training workers before workplace AI use becomes routine, or should employees carry the main responsibility for learning safe practices?
Sources and evidence
- Canadian workers are using AI faster than employers are training them – Digital Journal (10 November 2026, 23:09 UTC)
- What It Really Takes to Build an AI-Ready Workforce | HumanX 2026 Agenda | San Francisco | April 6-9, 2026 (10 November 2026, 23:09 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.