Hello {{first_name | Reader}},
Most of the risk in using AI at work is not in what you type into it. It is in what you connect it to.
✍️New this week: How to connect Gmail to ChatGPT safely?
Connecting your inbox to an AI assistant takes about four seconds. One screen, one button, done. What that button hands over is closer to a house key than a password: a pass that keeps working after you close the tab, until you go and remove it.
In 2025, attackers stole exactly that kind of pass from a sales tool and used it to read company email. Google was never hacked. The connection was the way in.
The post is the five-step setup I use, including the one setting that stops your assistant from sending anything without asking you first. Ten minutes, no technical knowledge needed.
📰 Story worth watching
On 16 September, Deloitte UK published the largest survey of workplace AI run in a single country: 25,000 workers, fielded by Ipsos. 63% use GenAI for work. Nearly a third of those users do it without their employer knowing.
The number that matters more sits further down. Around half of the people using AI at work have had no formal training on how to use it safely.
So the tools are already inside the building, connected to real inboxes and real documents, mostly by people who worked it out themselves. If that is you, the ten minutes above is the cheapest insurance you will buy this month.
That's it for this week.
If you found this useful, forward ⏩⏩ it to a colleague trying to figure out how AI agents work or reply to this email — I read everything.
See you soon, Linda