In the 1980s, long-distance phone companies had a trick called slamming.

You’d answer the phone, say “yes” to what sounded like an innocent question, and a month later your bill would show a new carrier you never chose. You didn’t opt in. You were switched. Quietly. The discovery came after the damage.

Regulators eventually called it what it was: fraud.

This week, Slack reminded me of that word. This application – which is an excellent chat system – has not yet been approved for AI use in our company.

It used to be that when I fell behind in a channel, a simple purple button said how many messages I missed.  Click it, and I’d jump to the first unread message. Clean. Deterministic. My choice.

Now the same purple button says, “Summarize messages.” I almost pressed it, out of habit, which would’ve caused me to violate our policy.

Nothing broke. Nothing crashed.

The feature probably helps someone. But the workflow changed without me asking. The default quietly nudged me toward AI.

That’s not assistance. That’s steering.

And Slack is only following the herd.  We are all being nudged towards a technology that has not yet been properly controlled. And I understand our government has decided not to regulate the providers of AI.

I’m not anti-AI. I use it every day.

But I want to choose when intelligence is augmented, and I want to understand the risk.

Replacing a neutral control with an AI shortcut feels less like a tool, and more like a shove.

The pattern is familiar. First, it’s optional. Then it’s prominent. Then it’s assumed. Eventually it’s unavoidable.

We’ve seen this movie before: pre-checked boxes, hidden fees, auto-renewals. Free apps that now cost us thousands of dollars per year to use. Dark patterns dressed up as convenience.

Now it’s intelligence itself being auto-enrolled.

Call it what it is: AI Slamming.

Switching the user’s process without explicit consent. Good tools respect agency. They wait to be invited. They don’t quietly move the furniture while you’re asleep.

If AI is really that good, it shouldn’t need to trick us into using it. And because good applications are now doing that, I believe our awareness training needs to introduce a new term: AI Slamming.