As businesses continue to infuse Ai assistance into every-day operations, it’s important that they stay diligent about what information they feed said programs.

That’s from Chris O’Sullivan, the Chief Information Officer at BFL CANADA. He says according to IBM, about 13 per cent of businesses globally experienced an Ai data exposure incident in 2025, which doubled year-over-year.

“Stats Canada put out a study around cyber incidents that shows about 1 in 6 companies having cyber security incidents, and we know Ai-related incidents are a part of that.”

He says this is no fault of the companies; the adoption of Ai is just happening too rapidly for them to keep up with all of the precautions that need to be put in place.

“The adoption is happening at such a pace that it’s really difficult for companies to keep up despite their best efforts. What you see is the adoption of these Ai technologies has happened with probably one of the fastest adoption rates of any technology.”

He adds that free Ai programs aren’t great for data protection and often require additional steps in order for someone to get important information back once submitted.

His tips for Ai safety include providing employee guardrails, providing training on how to use these tools securely, using enterprise versions of the tools, and monitoring potentially risky employee behavior.

Despite all this, O’Sullivan thinks “the potential for AI in terms of productivity and economic growth is tremendous. Is it worth it? Does it outweigh the risk of the clean up? Absolutely, I’d go there. The risk isn’t Ai, necessarily. It’s unmanaged Ai.”