Companies Don’t Adopt AI. People Do.
Getting people to want to do things 101.
75% of executives say their company’s AI strategy is more for show.
Somewhere, a leadership team is mapping out their AI strategy.
There’s a slide deck (always and forever).
The slide deck has phases:
Phase one is a tooling rollout. Copilot.
Phase two is training. How? Vendors. Someone “knows someone.”
Phase three is “AI fluency by Q3.” Read: Efficiency. Gains.
Phase four is, presumably, AI transcendence.
The corporate version of I bought this but I don’t wear it has been the narrative counterweight to bombastic AI financing rounds and fun tickets spent.
On my own LinkedIn profile, I note “I want to talk about your AI adoption problems,” and I do.
Efficiency gains and AI transcendence are not plug and play situations. Tech adoption, product adoption, service adoption. These are Behavioral Things.
Duh, well at least a retrospective duh, if that feels better.
The time-honored challenge stands: how do we get humans to do things?
More specifically (and appropriately), how do we get humans to want to do things?
And passive training won’t work. I’m experiencing this every minute of every day as I putter away on a conversational diagnostic that helps you identify patterns for AI in your personal and professional life.
I’m tripping over different things every day, things that aren’t called out on podcasts or YouTube videos or in Reddit threads or LinkedIn posts.
The question I’ve been percolating on for a long time now when it comes to AI, is how do we use curiosity, delight, and quick wins to overcome the “more work before less work” element of learning how to use a new tool that will ultimately help you?
Employees know they have access to tools like Copilot, and they’re aware that these tools are able to help them complete tasks more efficiently (though buy-in to this vision varies by experience and sentiment).
Out-Of-The-Box Use Cases like document summarization and proactively generated meeting notes encourage some but irritate most, with the latter sighing great… thanks.
More importantly, these now run-of-the-mill I Guess That’s Fine Use Cases don’t translate into someone showing up Tuesday having taken their (nonexistent) bandwidth to create a detailed automation spec.
Workers are getting stuck in:
the delta between their expectations and results from AI to date (significant)
the bandwidth they’re given by their employers to do additional learning on top of their day jobs (none)
and their ability to bridge the conceptual knowledge to mechanical application chasm (hi)
Going back to my point above, how do we create quick (low bandwidth needed) wins and momentum for workers to explore and engage? Said another way: how do we lower the first step on the staircase?
In my mind, the easiest way is through conversations. Don’t make workers square the circle on their own, the conceptual to mechanical gap; so many people are already stressed and tired.
Behavior change comes from activation energy, small first steps, and quick feedback loops.
Conversation creates an easy onramp to spot personalized, high value AI use cases that keep employees coming back for more once they’ve done a rep or two.
While curriculums and classes might work for some people, this is the entry point I can’t seem to sleep on.